Merge remote-tracking branch 'external/develop' into sync-develop-from-external

.azuredevops/components/ROCmValidationSuite.yml

@@ -86,8 +86,7 @@ jobs:
       value: $(Agent.BuildDirectory)/rocm
     - name: HIP_INC_DIR
       value: $(Agent.BuildDirectory)/rocm
-    pool:
+    pool: ${{ variables.MEDIUM_BUILD_POOL }}
-      vmImage: ${{ variables.BASE_BUILD_POOL }}
     workspace:
       clean: all
     steps:

.azuredevops/components/Tensile.yml

@@ -33,8 +33,9 @@ parameters:
   type: object
   default:
     - cmake
-    - libmsgpack-dev
+    - libboost-filesystem-dev
     - libboost-program-options-dev
+    - libmsgpack-dev
 - name: pipModules
   type: object
   default:

.azuredevops/components/composable_kernel.yml

@@ -107,6 +107,7 @@ jobs:
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/manifest.yml
       parameters:
         gpuTarget: ${{ job.target }}
+    # if this artifact name is changed, please also update $ARTIFACT_URL inside miopen-get-ck-build.yml
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/artifact-upload.yml
       parameters:
         gpuTarget: ${{ job.target }}

.azuredevops/components/copyHIP.yml

@@ -39,4 +39,6 @@ jobs:
       parameters:
         os: ${{ job.os }}
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/artifact-upload.yml
+      inputs:
+        os: ${{ job.os }}
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/artifact-links.yml

.azuredevops/components/hipBLAS-common.yml

@@ -51,15 +51,15 @@ parameters:
     buildJobs:
       - { os: ubuntu2204, packageManager: apt }
       - { os: almalinux8, packageManager: dnf }
-# - name: downstreamComponentMatrix
+- name: downstreamComponentMatrix
-#   type: object
+  type: object
-#   default:
+  default:
-#     - hipBLASLt:
+    - hipBLASLt:
-#       name: hipBLASLt
+      name: hipBLASLt
-#       sparseCheckoutDir: projects/hipblaslt
+      sparseCheckoutDir: projects/hipblaslt
-#       skipUnifiedBuild: 'false'
+      skipUnifiedBuild: 'false'
-#       buildDependsOn:
+      buildDependsOn:
-#         - hipBLAS_common_build
+        - hipBLAS_common_build
 
 jobs:
 - ${{ each job in parameters.jobMatrix.buildJobs }}:
@@ -122,14 +122,14 @@ jobs:
     #     extraEnvVars:
     #       - ROCM_PATH:::/home/user/workspace/rocm
 
-# - ${{ if parameters.triggerDownstreamJobs }}:
+- ${{ if parameters.triggerDownstreamJobs }}:
-#   - ${{ each component in parameters.downstreamComponentMatrix }}:
+  - ${{ each component in parameters.downstreamComponentMatrix }}:
-#     - ${{ if not(and(parameters.unifiedBuild, eq(component.skipUnifiedBuild, 'true'))) }}:
+    - ${{ if not(and(parameters.unifiedBuild, eq(component.skipUnifiedBuild, 'true'))) }}:
-#       - template: /.azuredevops/components/${{ component.name }}.yml@pipelines_repo
+      - template: /.azuredevops/components/${{ component.name }}.yml@pipelines_repo
-#         parameters:
+        parameters:
-#           checkoutRepo: ${{ parameters.checkoutRepo }}
+          checkoutRepo: ${{ parameters.checkoutRepo }}
-#           sparseCheckoutDir: ${{ component.sparseCheckoutDir }}
+          sparseCheckoutDir: ${{ component.sparseCheckoutDir }}
-#           buildDependsOn: ${{ component.buildDependsOn }}
+          buildDependsOn: ${{ component.buildDependsOn }}
-#           downstreamAggregateNames: ${{ parameters.downstreamAggregateNames }}+${{ parameters.componentName }}
+          downstreamAggregateNames: ${{ parameters.downstreamAggregateNames }}+${{ parameters.componentName }}
-#           triggerDownstreamJobs: true
+          triggerDownstreamJobs: true
-#           unifiedBuild: ${{ parameters.unifiedBuild }}
+          unifiedBuild: ${{ parameters.unifiedBuild }}

.azuredevops/components/hipBLASLt.yml

@@ -77,28 +77,28 @@ parameters:
   type: object
   default:
     buildJobs:
-      - { os: ubuntu2204, packageManager: apt, target: gfx942 }
+      - { pool: rocm-ci_ultra_build_pool, os: ubuntu2204, packageManager: apt, target: gfx942 }
-      - { os: ubuntu2204, packageManager: apt, target: gfx90a }
+      - { pool: rocm-ci_medium_build_pool, os: ubuntu2204, packageManager: apt, target: gfx90a }
-      - { os: ubuntu2204, packageManager: apt, target: gfx1201 }
+      - { pool: rocm-ci_medium_build_pool, os: ubuntu2204, packageManager: apt, target: gfx1201 }
-      # - { os: ubuntu2204, packageManager: apt, target: gfx1100 }
+      - { pool: rocm-ci_medium_build_pool, os: ubuntu2204, packageManager: apt, target: gfx1100 }
-      - { os: ubuntu2204, packageManager: apt, target: gfx1030 }
+      - { pool: rocm-ci_medium_build_pool, os: ubuntu2204, packageManager: apt, target: gfx1030 }
-      # - { os: almalinux8, packageManager: dnf, target: gfx942 }
+      - { pool: rocm-ci_ultra_build_pool, os: almalinux8, packageManager: dnf, target: gfx942 }
-      # - { os: almalinux8, packageManager: dnf, target: gfx90a }
+      - { pool: rocm-ci_medium_build_pool, os: almalinux8, packageManager: dnf, target: gfx90a }
-      # - { os: almalinux8, packageManager: dnf, target: gfx1201 }
+      - { pool: rocm-ci_medium_build_pool, os: almalinux8, packageManager: dnf, target: gfx1201 }
-      # - { os: almalinux8, packageManager: dnf, target: gfx1100 }
+      - { pool: rocm-ci_medium_build_pool, os: almalinux8, packageManager: dnf, target: gfx1100 }
-      # - { os: almalinux8, packageManager: dnf, target: gfx1030 }
+      - { pool: rocm-ci_medium_build_pool, os: almalinux8, packageManager: dnf, target: gfx1030 }
     testJobs:
       - { os: ubuntu2204, packageManager: apt, target: gfx942 }
       - { os: ubuntu2204, packageManager: apt, target: gfx90a }
-# - name: downstreamComponentMatrix
+- name: downstreamComponentMatrix
-#   type: object
+  type: object
-#   default:
+  default:
-#     - rocBLAS:
+    - rocBLAS:
-#       name: rocBLAS
+      name: rocBLAS
-#       sparseCheckoutDir: projects/rocblas
+      sparseCheckoutDir: projects/rocblas
-#       skipUnifiedBuild: 'false'
+      skipUnifiedBuild: 'false'
-#       buildDependsOn:
+      buildDependsOn:
-#         - hipBLASLt_build
+        - hipBLASLt_build
 
 jobs:
 - ${{ each job in parameters.jobMatrix.buildJobs }}:
@@ -121,7 +121,7 @@ jobs:
       value: $(Agent.BuildDirectory)/rocm
     - name: DAY_STRING
       value: $[format('{0:ddMMyyyy}', pipeline.startTime)]
-    pool: ${{ variables.ULTRA_BUILD_POOL }}
+    pool: ${{ job.pool }}
     ${{ if eq(job.os, 'almalinux8') }}:
       container:
         image: rocmexternalcicd.azurecr.io/manylinux228:latest
@@ -140,6 +140,10 @@ jobs:
       parameters:
         checkoutRepo: ${{ parameters.checkoutRepo }}
         sparseCheckoutDir: ${{ parameters.sparseCheckoutDir }}
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-vendor.yml
+      parameters:
+        dependencyList:
+          - gtest
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
       parameters:
         checkoutRef: ${{ parameters.checkoutRef }}
@@ -156,18 +160,15 @@ jobs:
         script: |
           echo "##vso[task.prependpath]$(Agent.BuildDirectory)/rocm/bin"
           echo "##vso[task.prependpath]$(Agent.BuildDirectory)/rocm/llvm/bin"
-    # hipBLASLt has a script for gtest and lapack
-    # https://github.com/ROCm/hipBLASLt/blob/develop/deps/CMakeLists.txt
-    # $(Agent.BuildDirectory)/deps is a temporary folder for the build process
-    # $(Agent.BuildDirectory)/s/deps is part of the hipBLASLt repo
     - task: Bash@3
-      displayName: Build and install external dependencies
+      displayName: Build and install LAPACK
       inputs:
         targetType: inline
         script: |
-          mkdir -p $(Agent.BuildDirectory)/deps
+          mkdir -p $(Agent.BuildDirectory)/temp-deps
-          cd $(Agent.BuildDirectory)/deps
+          cd $(Agent.BuildDirectory)/temp-deps
-          cmake -DCMAKE_POSITION_INDEPENDENT_CODE=ON $(Agent.BuildDirectory)/s/deps
+          # position-independent LAPACK is required for almalinux8 builds
+          cmake -DBUILD_GTEST=OFF -DBUILD_LAPACK=ON -DCMAKE_POSITION_INDEPENDENT_CODE=ON $(Agent.BuildDirectory)/s/deps
           make
           sudo make install
     - script: |
@@ -187,7 +188,7 @@ jobs:
       parameters:
         os: ${{ job.os }}
         extraBuildFlags: >-
-          -DCMAKE_PREFIX_PATH=$(Agent.BuildDirectory)/rocm
+          -DCMAKE_PREFIX_PATH=$(Agent.BuildDirectory)/rocm;$(Agent.BuildDirectory)/vendor
           -DCMAKE_INCLUDE_PATH=$(Agent.BuildDirectory)/rocm/llvm/include
           -DCMAKE_CXX_COMPILER=$(Agent.BuildDirectory)/rocm/llvm/bin/amdclang++
           -DCMAKE_C_COMPILER=$(Agent.BuildDirectory)/rocm/llvm/bin/amdclang
@@ -244,6 +245,7 @@ jobs:
       workspace:
         clean: all
       steps:
+      - checkout: none
       - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
         parameters:
           aptPackages: ${{ parameters.aptPackages }}
@@ -280,14 +282,14 @@ jobs:
           environment: test
           gpuTarget: ${{ job.target }}
 
-# - ${{ if parameters.triggerDownstreamJobs }}:
+- ${{ if parameters.triggerDownstreamJobs }}:
-#   - ${{ each component in parameters.downstreamComponentMatrix }}:
+  - ${{ each component in parameters.downstreamComponentMatrix }}:
-#     - ${{ if not(and(parameters.unifiedBuild, eq(component.skipUnifiedBuild, 'true'))) }}:
+    - ${{ if not(and(parameters.unifiedBuild, eq(component.skipUnifiedBuild, 'true'))) }}:
-#       - template: /.azuredevops/components/${{ component.name }}.yml@pipelines_repo
+      - template: /.azuredevops/components/${{ component.name }}.yml@pipelines_repo
-#         parameters:
+        parameters:
-#           checkoutRepo: ${{ parameters.checkoutRepo }}
+          checkoutRepo: ${{ parameters.checkoutRepo }}
-#           sparseCheckoutDir: ${{ component.sparseCheckoutDir }}
+          sparseCheckoutDir: ${{ component.sparseCheckoutDir }}
-#           buildDependsOn: ${{ component.buildDependsOn }}
+          buildDependsOn: ${{ component.buildDependsOn }}
-#           downstreamAggregateNames: ${{ parameters.downstreamAggregateNames }}+${{ parameters.componentName }}
+          downstreamAggregateNames: ${{ parameters.downstreamAggregateNames }}+${{ parameters.componentName }}
-#           triggerDownstreamJobs: true
+          triggerDownstreamJobs: true
-#           unifiedBuild: ${{ parameters.unifiedBuild }}
+          unifiedBuild: ${{ parameters.unifiedBuild }}

.azuredevops/components/hipFFT.yml

@@ -156,6 +156,7 @@ jobs:
     workspace:
       clean: all
     steps:
+    - checkout: none
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
       parameters:
         aptPackages: ${{ parameters.aptPackages }}

.azuredevops/components/hipSOLVER.yml

@@ -70,8 +70,7 @@ jobs:
     variables:
     - group: common
     - template: /.azuredevops/variables-global.yml
-    pool:
+    pool: ${{ variables.MEDIUM_BUILD_POOL }}
-      vmImage: ${{ variables.BASE_BUILD_POOL }}
     workspace:
       clean: all
     steps:

.azuredevops/components/hipSPARSE.yml

@@ -1,10 +1,29 @@
 parameters:
+- name: componentName
+  type: string
+  default: hipSPARSE
 - name: checkoutRepo
   type: string
   default: 'self'
 - name: checkoutRef
   type: string
   default: ''
+# monorepo related parameters
+- name: sparseCheckoutDir
+  type: string
+  default: ''
+- name: triggerDownstreamJobs
+  type: boolean
+  default: false
+- name: downstreamAggregateNames
+  type: string
+  default: ''
+- name: buildDependsOn
+  type: object
+  default: null
+- name: unifiedBuild
+  type: boolean
+  default: false
 # set to true if doing full build of ROCm stack
 # and dependencies are pulled from same pipeline
 - name: aggregatePipeline
@@ -14,13 +33,11 @@ parameters:
   type: object
   default:
     - cmake
-    - ninja-build
-    - libboost-program-options-dev
-    - googletest
-    - libfftw3-dev
-    - git
     - gfortran
-    - libgtest-dev
+    - git
+    - libboost-program-options-dev
+    - libfftw3-dev
+    - ninja-build
     - python3-pip
 - name: rocmDependencies
   type: object
@@ -49,19 +66,31 @@ parameters:
   type: object
   default:
     buildJobs:
-      - gfx942:
+      - { os: ubuntu2204, packageManager: apt, target: gfx942 }
-        target: gfx942
+      - { os: ubuntu2204, packageManager: apt, target: gfx90a }
-      - gfx90a:
+      - { os: ubuntu2204, packageManager: apt, target: gfx1201 }
-        target: gfx90a
+      - { os: ubuntu2204, packageManager: apt, target: gfx1030 }
+      - { os: ubuntu2204, packageManager: apt, target: gfx1100 }
     testJobs:
-      - gfx942:
+      - { os: ubuntu2204, packageManager: apt, target: gfx942 }
-        target: gfx942
+      - { os: ubuntu2204, packageManager: apt, target: gfx90a }
-      - gfx90a:
+- name: downstreamComponentMatrix
-        target: gfx90a
+  type: object
+  default:
+    - hipSPARSELt:
+      name: hipSPARSELt
+      sparseCheckoutDir: projects/hipsparselt
+      skipUnifiedBuild: 'false'
+      buildDependsOn:
+        - hipSPARSE_build
 
 jobs:
 - ${{ each job in parameters.jobMatrix.buildJobs }}:
-  - job: hipSPARSE_build_${{ job.target }}
+  - job: ${{ parameters.componentName }}_build_${{ job.os }}_${{ job.target }}
+    ${{ if parameters.buildDependsOn }}:
+      dependsOn:
+        - ${{ each build in parameters.buildDependsOn }}:
+          - ${{ build }}_${{ job.os }}_${{ job.target }}
     variables:
     - group: common
     - template: /.azuredevops/variables-global.yml
@@ -73,42 +102,57 @@ jobs:
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
       parameters:
         aptPackages: ${{ parameters.aptPackages }}
+        packageManager: ${{ job.packageManager }}
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/preamble.yml
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/checkout.yml
       parameters:
         checkoutRepo: ${{ parameters.checkoutRepo }}
+        sparseCheckoutDir: ${{ parameters.sparseCheckoutDir }}
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-vendor.yml
+      parameters:
+        dependencyList:
+          - gtest
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
       parameters:
         checkoutRef: ${{ parameters.checkoutRef }}
         dependencyList: ${{ parameters.rocmDependencies }}
         gpuTarget: ${{ job.target }}
+        os: ${{ job.os }}
         aggregatePipeline: ${{ parameters.aggregatePipeline }}
+        ${{ if parameters.triggerDownstreamJobs }}:
+          downstreamAggregateNames: ${{ parameters.downstreamAggregateNames }}
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
       parameters:
+        os: ${{ job.os }}
         extraBuildFlags: >-
+          -DCMAKE_PREFIX_PATH=$(Agent.BuildDirectory)/rocm;$(Agent.BuildDirectory)/vendor
           -DCMAKE_CXX_COMPILER=$(Agent.BuildDirectory)/rocm/llvm/bin/amdclang++
+          -DCMAKE_C_COMPILER=$(Agent.BuildDirectory)/rocm/bin/amdclang
           -DCMAKE_BUILD_TYPE=Release
-          -DCMAKE_PREFIX_PATH=$(Agent.BuildDirectory)/rocm;$(Agent.BuildDirectory)/rocm/share/rocm/cmake/
           -DBUILD_CLIENTS_TESTS=ON
           -DBUILD_CLIENTS_SAMPLES=OFF
           -GNinja
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/manifest.yml
       parameters:
-        artifactName: hipSPARSE
+        componentName: ${{ parameters.componentName }}
+        sparseCheckoutDir: ${{ parameters.sparseCheckoutDir }}
         gpuTarget: ${{ job.target }}
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/artifact-upload.yml
       parameters:
-        artifactName: hipSPARSE
+        componentName: ${{ parameters.componentName }}
         gpuTarget: ${{ job.target }}
+        os: ${{ job.os }}
         publish: false
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/artifact-prepare-package.yml
       parameters:
-        sourceDir: $(Build.SourcesDirectory)/build/clients
+        sourceDir: $(Agent.BuildDirectory)/s/build/clients
         contentsString: matrices/**
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/artifact-upload.yml
       parameters:
+        componentName: ${{ parameters.componentName }}
         artifactName: testMatrices
         gpuTarget: ${{ job.target }}
+        os: ${{ job.os }}
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/artifact-links.yml
     # - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/docker-container.yml
     #   parameters:
@@ -116,44 +160,65 @@ jobs:
     #     environment: test
     #     gpuTarget: ${{ job.target }}
 
-- ${{ each job in parameters.jobMatrix.testJobs }}:
+- ${{ if eq(parameters.unifiedBuild, False) }}:
-  - job: hipSPARSE_test_${{ job.target }}
+  - ${{ each job in parameters.jobMatrix.testJobs }}:
-    dependsOn: hipSPARSE_build_${{ job.target }}
+    - job: ${{ parameters.componentName }}_test_${{ job.os }}_${{ job.target }}
-    condition:
+      dependsOn: ${{ parameters.componentName }}_build_${{ job.os }}_${{ job.target }}
-      and(succeeded(),
+      condition:
-        eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
+        and(succeeded(),
-        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName'])),
+          eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
-        eq(${{ parameters.aggregatePipeline }}, False)
+          not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), '${{ parameters.componentName }}')),
-      )
+          eq(${{ parameters.aggregatePipeline }}, False)
-    variables:
+        )
-    - group: common
+      variables:
-    - template: /.azuredevops/variables-global.yml
+      - group: common
-    pool: ${{ job.target }}_test_pool
+      - template: /.azuredevops/variables-global.yml
-    workspace:
+      pool: ${{ job.target }}_test_pool
-      clean: all
+      workspace:
-    steps:
+        clean: all
-    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
+      steps:
-      parameters:
+      - checkout: none
-        aptPackages: ${{ parameters.aptPackages }}
+      - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
-    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/preamble.yml
+        parameters:
-    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/local-artifact-download.yml
+          aptPackages: ${{ parameters.aptPackages }}
-      parameters:
+          packageManager: ${{ job.packageManager }}
-        gpuTarget: ${{ job.target }}
+      - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/preamble.yml
-    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-aqlprofile.yml
+      - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/local-artifact-download.yml
-    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+        parameters:
-      parameters:
+          preTargetFilter: ${{ parameters.componentName }}
-        checkoutRef: ${{ parameters.checkoutRef }}
+          gpuTarget: ${{ job.target }}
-        dependencyList: ${{ parameters.rocmTestDependencies }}
+          os: ${{ job.os }}
-        gpuTarget: ${{ job.target }}
+      - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-aqlprofile.yml
-    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/gpu-diagnostics.yml
+      - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
-    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/test.yml
+        parameters:
-      parameters:
+          checkoutRef: ${{ parameters.checkoutRef }}
-        componentName: hipSPARSE
+          dependencyList: ${{ parameters.rocmTestDependencies }}
-        testDir: '$(Agent.BuildDirectory)/rocm/bin'
+          gpuTarget: ${{ job.target }}
-        testExecutable: './hipsparse-test'
+          os: ${{ job.os }}
-        testParameters: '--gtest_output=xml:./test_output.xml --gtest_color=yes'
+          ${{ if parameters.triggerDownstreamJobs }}:
-    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/docker-container.yml
+            downstreamAggregateNames: ${{ parameters.downstreamAggregateNames }}
-      parameters:
+      - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/gpu-diagnostics.yml
-        aptPackages: ${{ parameters.aptPackages }}
+      - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/test.yml
-        environment: test
+        parameters:
-        gpuTarget: ${{ job.target }}
+          componentName: ${{ parameters.componentName }}
+          os: ${{ job.os }}
+          testDir: '$(Agent.BuildDirectory)/rocm/bin'
+          testExecutable: './hipsparse-test'
+          testParameters: '--gtest_output=xml:./test_output.xml --gtest_color=yes'
+      - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/docker-container.yml
+        parameters:
+          aptPackages: ${{ parameters.aptPackages }}
+          environment: test
+          gpuTarget: ${{ job.target }}
+
+  - ${{ if parameters.triggerDownstreamJobs }}:
+    - ${{ each component in parameters.downstreamComponentMatrix }}:
+      - ${{ if not(and(parameters.unifiedBuild, eq(component.skipUnifiedBuild, 'true'))) }}:
+        - template: /.azuredevops/components/${{ component.name }}.yml@pipelines_repo
+          parameters:
+            checkoutRepo: ${{ parameters.checkoutRepo }}
+            sparseCheckoutDir: ${{ component.sparseCheckoutDir }}
+            buildDependsOn: ${{ component.buildDependsOn }}
+            downstreamAggregateNames: ${{ parameters.downstreamAggregateNames }}+${{ parameters.componentName }}
+            triggerDownstreamJobs: true
+            unifiedBuild: ${{ parameters.unifiedBuild }}

.azuredevops/components/hipSPARSELt.yml

@@ -1,10 +1,29 @@
 parameters:
+- name: componentName
+  type: string
+  default: hipSPARSELt
 - name: checkoutRepo
   type: string
   default: 'self'
 - name: checkoutRef
   type: string
   default: ''
+# monorepo related parameters
+- name: sparseCheckoutDir
+  type: string
+  default: ''
+- name: triggerDownstreamJobs
+  type: boolean
+  default: false
+- name: downstreamAggregateNames
+  type: string
+  default: ''
+- name: buildDependsOn
+  type: object
+  default: null
+- name: unifiedBuild
+  type: boolean
+  default: false
 # set to true if doing full build of ROCm stack
 # and dependencies are pulled from same pipeline
 - name: aggregatePipeline
@@ -56,15 +75,17 @@ parameters:
   type: object
   default:
     buildJobs:
-      - gfx942:
+      - { os: ubuntu2204, packageManager: apt, target: gfx942 }
-        target: gfx942
     testJobs:
-      - gfx942:
+      - { os: ubuntu2204, packageManager: apt, target: gfx942 }
-        target: gfx942
 
 jobs:
 - ${{ each job in parameters.jobMatrix.buildJobs }}:
-  - job: hipSPARSELt_build_${{ job.target }}
+  - job: ${{ parameters.componentName }}_build_${{ job.os }}_${{ job.target }}
+    ${{ if parameters.buildDependsOn }}:
+      dependsOn:
+        - ${{ each build in parameters.buildDependsOn }}:
+          - ${{ build }}_${{ job.os }}_${{ job.target }}
     variables:
     - group: common
     - template: /.azuredevops/variables-global.yml
@@ -86,17 +107,22 @@ jobs:
       parameters:
         aptPackages: ${{ parameters.aptPackages }}
         pipModules: ${{ parameters.pipModules }}
+        packageManager: ${{ job.packageManager }}
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-cmake-latest.yml
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/preamble.yml
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/checkout.yml
       parameters:
         checkoutRepo: ${{ parameters.checkoutRepo }}
+        sparseCheckoutDir: ${{ parameters.sparseCheckoutDir }}
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
       parameters:
         checkoutRef: ${{ parameters.checkoutRef }}
         dependencyList: ${{ parameters.rocmDependencies }}
         gpuTarget: ${{ job.target }}
+        os: ${{ job.os }}
         aggregatePipeline: ${{ parameters.aggregatePipeline }}
+        ${{ if parameters.triggerDownstreamJobs }}:
+          downstreamAggregateNames: ${{ parameters.downstreamAggregateNames }}
   # Build and install gtest and lapack
   # $(Pipeline.Workspace)/deps is a temporary folder for the build process
   # $(Pipeline.Workspace)/s/deps is part of the hipSPARSELt repo
@@ -115,6 +141,7 @@ jobs:
       workingDirectory: $(Pipeline.Workspace)/deps
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
       parameters:
+        os: ${{ job.os }}
         extraBuildFlags: >-
           -DCMAKE_BUILD_TYPE=Release
           -DCMAKE_CXX_COMPILER=$(Agent.BuildDirectory)/rocm/llvm/bin/amdclang++
@@ -130,64 +157,80 @@ jobs:
           -GNinja
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/manifest.yml
       parameters:
+        componentName: ${{ parameters.componentName }}
         gpuTarget: ${{ job.target }}
+        os: ${{ job.os }}
+        sparseCheckoutDir: ${{ parameters.sparseCheckoutDir }}
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/artifact-upload.yml
       parameters:
+        componentName: ${{ parameters.componentName }}
         gpuTarget: ${{ job.target }}
+        os: ${{ job.os }}
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/artifact-links.yml
-    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/docker-container.yml
+    - ${{ if eq(job.os, 'ubuntu2204') }}:
-      parameters:
+      - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/docker-container.yml
-        aptPackages: ${{ parameters.aptPackages }}
+        parameters:
-        pipModules: ${{ parameters.pipModules }}
+          aptPackages: ${{ parameters.aptPackages }}
-        gpuTarget: ${{ job.target }}
+          pipModules: ${{ parameters.pipModules }}
-        extraCopyDirectories:
+          gpuTarget: ${{ job.target }}
-          - deps
+          extraCopyDirectories:
-        extraPaths: /home/user/workspace/rocm/llvm/bin:/home/user/workspace/rocm/bin
+            - deps
-        extraEnvVars:
+          extraPaths: /home/user/workspace/rocm/llvm/bin:/home/user/workspace/rocm/bin
-          - HIP_ROCCLR_HOME:::/home/user/workspace/rocm
+          extraEnvVars:
-          - TENSILE_ROCM_ASSEMBLER_PATH:::/home/user/workspace/rocm/llvm/bin/clang
+            - HIP_ROCCLR_HOME:::/home/user/workspace/rocm
-          - CMAKE_CXX_COMPILER:::/home/user/workspace/rocm/llvm/bin/hipcc
+            - TENSILE_ROCM_ASSEMBLER_PATH:::/home/user/workspace/rocm/llvm/bin/clang
-          - TENSILE_ROCM_OFFLOAD_BUNDLER_PATH:::/home/user/workspace/rocm/llvm/bin/clang-offload-bundler
+            - CMAKE_CXX_COMPILER:::/home/user/workspace/rocm/llvm/bin/hipcc
-        installLatestCMake: true
+            - TENSILE_ROCM_OFFLOAD_BUNDLER_PATH:::/home/user/workspace/rocm/llvm/bin/clang-offload-bundler
+          installLatestCMake: true
 
-- ${{ each job in parameters.jobMatrix.testJobs }}:
+- ${{ if eq(parameters.unifiedBuild, False) }}:
-  - job: hipSPARSELt_test_${{ job.target }}
+  - ${{ each job in parameters.jobMatrix.testJobs }}:
-    dependsOn: hipSPARSELt_build_${{ job.target }}
+    - job: ${{ parameters.componentName }}_test_${{ job.os }}_${{ job.target }}
-    condition:
+      timeoutInMinutes: 120
-      and(succeeded(),
+      dependsOn: ${{ parameters.componentName }}_build_${{ job.os }}_${{ job.target }}
-        eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
+      condition:
-        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName'])),
+        and(succeeded(),
-        eq(${{ parameters.aggregatePipeline }}, False)
+          eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
-      )
+          not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), '${{ parameters.componentName }}')),
-    variables:
+          eq(${{ parameters.aggregatePipeline }}, False)
-    - group: common
+        )
-    - template: /.azuredevops/variables-global.yml
+      variables:
-    pool: ${{ job.target }}_test_pool
+      - group: common
-    workspace:
+      - template: /.azuredevops/variables-global.yml
-      clean: all
+      pool: ${{ job.target }}_test_pool
-    steps:
+      workspace:
-    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
+        clean: all
-      parameters:
+      steps:
-        aptPackages: ${{ parameters.aptPackages }}
+      - checkout: none
-    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/preamble.yml
+      - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
-    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/local-artifact-download.yml
+        parameters:
-      parameters:
+          aptPackages: ${{ parameters.aptPackages }}
-        gpuTarget: ${{ job.target }}
+          packageManager: ${{ job.packageManager }}
-    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/preamble.yml
-      parameters:
+      - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/local-artifact-download.yml
-        checkoutRef: ${{ parameters.checkoutRef }}
+        parameters:
-        dependencyList: ${{ parameters.rocmTestDependencies }}
+          preTargetFilter: ${{ parameters.componentName }}
-        gpuTarget: ${{ job.target }}
+          gpuTarget: ${{ job.target }}
-    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/gpu-diagnostics.yml
+          os: ${{ job.os }}
-    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/test.yml
+      - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
-      parameters:
+        parameters:
-        componentName: hipSPARSELt
+          checkoutRef: ${{ parameters.checkoutRef }}
-        testDir: '$(Agent.BuildDirectory)/rocm/bin'
+          dependencyList: ${{ parameters.rocmTestDependencies }}
-        testExecutable: './hipsparselt-test'
+          gpuTarget: ${{ job.target }}
-        testParameters: '--gtest_output=xml:./test_output.xml --gtest_color=yes --gtest_filter=*pre_checkin*'
+          os: ${{ job.os }}
-    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/docker-container.yml
+          ${{ if parameters.triggerDownstreamJobs }}:
-      parameters:
+            downstreamAggregateNames: ${{ parameters.downstreamAggregateNames }}
-        aptPackages: ${{ parameters.aptPackages }}
+      - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/gpu-diagnostics.yml
-        pipModules: ${{ parameters.pipModules }}
+      - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/test.yml
-        environment: test
+        parameters:
-        gpuTarget: ${{ job.target }}
+          componentName: ${{ parameters.componentName }}
+          os: ${{ job.os }}
+          testDir: '$(Agent.BuildDirectory)/rocm/bin'
+          testExecutable: './hipsparselt-test'
+          testParameters: '--gtest_output=xml:./test_output.xml --gtest_color=yes --gtest_filter=*pre_checkin*'
+      - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/docker-container.yml
+        parameters:
+          aptPackages: ${{ parameters.aptPackages }}
+          pipModules: ${{ parameters.pipModules }}
+          environment: test
+          gpuTarget: ${{ job.target }}

.azuredevops/components/llvm-project.yml

@@ -67,7 +67,6 @@ jobs:
       parameters:
         checkoutRef: ${{ parameters.checkoutRef }}
         dependencyList: ${{ parameters.rocmDependencies }}
-        skipLlvmSymlink: true
         aggregatePipeline: ${{ parameters.aggregatePipeline }}
         os: ${{ job.os }}
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml

.azuredevops/components/rocAL.yml

@@ -86,8 +86,7 @@ jobs:
     variables:
     - group: common
     - template: /.azuredevops/variables-global.yml
-    pool:
+    pool:  ${{ variables.MEDIUM_BUILD_POOL }}
-      vmImage: ${{ variables.BASE_BUILD_POOL }}
     workspace:
       clean: all
     steps:

.azuredevops/components/rocALUTION.yml

@@ -73,8 +73,7 @@ jobs:
     - template: /.azuredevops/variables-global.yml
     - name: HIP_ROCCLR_HOME
       value: $(Build.BinariesDirectory)/rocm
-    pool:
+    pool: ${{ variables.MEDIUM_BUILD_POOL }}
-      vmImage: ${{ variables.BASE_BUILD_POOL }}
     workspace:
       clean: all
     steps:

.azuredevops/components/rocBLAS.yml

@@ -33,17 +33,15 @@ parameters:
   type: object
   default:
     - cmake
-    - ninja-build
-    - python3-venv
     - git
-    - libmsgpack-dev
     - gfortran
-    - libopenblas-dev
-    - googletest
-    - libgtest-dev
-    - wget
-    - python3-pip
     - libdrm-dev
+    - libmsgpack-dev
+    - libopenblas-dev
+    - ninja-build
+    - python3-pip
+    - python3-venv
+    - wget
 - name: pipModules
   type: object
   default:
@@ -52,18 +50,17 @@ parameters:
 - name: rocmDependencies
   type: object
   default:
-    - rocm-cmake
-    - llvm-project
-    - ROCR-Runtime
-    - clr
-    - rocminfo
-    - rocprofiler-register
-    - rocm_smi_lib
-    - rocm-core
     - aomp
-    - aomp-extras
+    - clr
     - hipBLAS-common
     - hipBLASLt
+    - llvm-project
+    - rocm-cmake
+    - rocm-core
+    - rocm_smi_lib
+    - rocminfo
+    - rocprofiler-register
+    - ROCR-Runtime
     - roctracer
 - name: rocmTestDependencies
   type: object
@@ -86,32 +83,38 @@ parameters:
       - { os: ubuntu2204, packageManager: apt, target: gfx942 }
       - { os: ubuntu2204, packageManager: apt, target: gfx90a }
       - { os: ubuntu2204, packageManager: apt, target: gfx1201 }
-      # - { os: ubuntu2204, packageManager: apt, target: gfx1100 }
+      - { os: ubuntu2204, packageManager: apt, target: gfx1100 }
       - { os: ubuntu2204, packageManager: apt, target: gfx1030 }
-      # - { os: almalinux8, packageManager: dnf, target: gfx942 }
+      - { os: almalinux8, packageManager: dnf, target: gfx942 }
-      # - { os: almalinux8, packageManager: dnf, target: gfx90a }
+      - { os: almalinux8, packageManager: dnf, target: gfx90a }
-      # - { os: almalinux8, packageManager: dnf, target: gfx1201 }
+      - { os: almalinux8, packageManager: dnf, target: gfx1201 }
-      # - { os: almalinux8, packageManager: dnf, target: gfx1100 }
+      - { os: almalinux8, packageManager: dnf, target: gfx1100 }
-      # - { os: almalinux8, packageManager: dnf, target: gfx1030 }
+      - { os: almalinux8, packageManager: dnf, target: gfx1030 }
     testJobs:
       - { os: ubuntu2204, packageManager: apt, target: gfx942 }
       - { os: ubuntu2204, packageManager: apt, target: gfx90a }
-# - name: downstreamComponentMatrix
+- name: downstreamComponentMatrix
-#   type: object
+  type: object
-#   default:
+  default:
-#     # rocSOLVER depends on both rocBLAS and rocPRIM
+    - rocSPARSE:
-#     # for a unified build, rocBLAS will be the one to call rocSOLVER
+      name: rocSPARSE
-#     - rocSOLVER:
+      sparseCheckoutDir: projects/rocsparse
-#       name: rocSOLVER
+      skipUnifiedBuild: 'false'
-#       sparseCheckoutDir: projects/rocsolver
+      buildDependsOn:
-#       skipUnifiedBuild: 'false'
+        - rocBLAS_build
-#       buildDependsOn:
+    # rocSOLVER depends on both rocBLAS and rocPRIM
-#         - rocBLAS_build
+    # for a unified build, rocBLAS will be the one to call rocSOLVER
-#       unifiedBuild:
+    # - rocSOLVER:
-#         downstreamAggregateNames: rocBLAS+rocPRIM
+    #   name: rocSOLVER
-#         buildDependsOn:
+    #   sparseCheckoutDir: projects/rocsolver
-#           - rocBLAS_build
+    #   skipUnifiedBuild: 'false'
-#           - rocPRIM_build
+    #   buildDependsOn:
+    #     - rocBLAS_build
+    #   unifiedBuild:
+    #     downstreamAggregateNames: rocBLAS+rocPRIM
+    #     buildDependsOn:
+    #       - rocBLAS_build
+    #       - rocPRIM_build
 
 jobs:
 - ${{ each job in parameters.jobMatrix.buildJobs }}:
@@ -151,6 +154,12 @@ jobs:
         checkoutRepo: ${{ parameters.checkoutRepo }}
         sparseCheckoutDir: ${{ parameters.sparseCheckoutDir }}
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-aocl.yml
+      parameters:
+        os: ${{ job.os }}
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-vendor.yml
+      parameters:
+        dependencyList:
+          - gtest
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
       parameters:
         checkoutRef: ${{ parameters.checkoutRef }}
@@ -164,21 +173,12 @@ jobs:
       parameters:
         os: ${{ job.os }}
         extraBuildFlags: >-
-          -DCMAKE_TOOLCHAIN_FILE=toolchain-linux.cmake
+          -DCMAKE_PREFIX_PATH=$(Agent.BuildDirectory)/rocm/llvm;$(Agent.BuildDirectory)/rocm;$(Agent.BuildDirectory)/vendor
-          -DCMAKE_PREFIX_PATH=$(Agent.BuildDirectory)/rocm/llvm;$(Agent.BuildDirectory)/rocm
           -DCMAKE_BUILD_TYPE=Release
           -DCMAKE_CXX_COMPILER=$(Agent.BuildDirectory)/rocm/bin/amdclang++
           -DCMAKE_C_COMPILER=$(Agent.BuildDirectory)/rocm/bin/amdclang
           -DGPU_TARGETS=${{ job.target }}
-          -DTensile_CODE_OBJECT_VERSION=default
-          -DTensile_LOGIC=asm_full
-          -DTensile_SEPARATE_ARCHITECTURES=ON
-          -DTensile_LAZY_LIBRARY_LOADING=ON
-          -DTensile_LIBRARY_FORMAT=msgpack
           -DBUILD_CLIENTS_TESTS=ON
-          -DBUILD_CLIENTS_BENCHMARKS=OFF
-          -DBUILD_CLIENTS_SAMPLES=OFF
-          -DROCM_PATH=$(Agent.BuildDirectory)/rocm
           -GNinja
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/manifest.yml
       parameters:
@@ -208,6 +208,7 @@ jobs:
 - ${{ if eq(parameters.unifiedBuild, False) }}:
   - ${{ each job in parameters.jobMatrix.testJobs }}:
     - job: ${{ parameters.componentName }}_test_${{ job.os }}_${{ job.target }}
+      timeoutInMinutes: 120
       dependsOn: ${{ parameters.componentName }}_build_${{ job.os }}_${{ job.target }}
       condition:
         and(succeeded(),
@@ -222,6 +223,7 @@ jobs:
       workspace:
         clean: all
       steps:
+      - checkout: none
       - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
         parameters:
           aptPackages: ${{ parameters.aptPackages }}
@@ -258,18 +260,18 @@ jobs:
           environment: test
           gpuTarget: ${{ job.target }}
 
-# - ${{ if parameters.triggerDownstreamJobs }}:
+- ${{ if parameters.triggerDownstreamJobs }}:
-#   - ${{ each component in parameters.downstreamComponentMatrix }}:
+  - ${{ each component in parameters.downstreamComponentMatrix }}:
-#     - ${{ if not(and(parameters.unifiedBuild, eq(component.skipUnifiedBuild, 'true'))) }}:
+    - ${{ if not(and(parameters.unifiedBuild, eq(component.skipUnifiedBuild, 'true'))) }}:
-#       - template: /.azuredevops/components/${{ component.name }}.yml@pipelines_repo
+      - template: /.azuredevops/components/${{ component.name }}.yml@pipelines_repo
-#         parameters:
+        parameters:
-#           checkoutRepo: ${{ parameters.checkoutRepo }}
+          checkoutRepo: ${{ parameters.checkoutRepo }}
-#           sparseCheckoutDir: ${{ component.sparseCheckoutDir }}
+          sparseCheckoutDir: ${{ component.sparseCheckoutDir }}
-#           triggerDownstreamJobs: true
+          triggerDownstreamJobs: true
-#           unifiedBuild: ${{ parameters.unifiedBuild }}
+          unifiedBuild: ${{ parameters.unifiedBuild }}
-#           ${{ if parameters.unifiedBuild }}:
+          ${{ if parameters.unifiedBuild }}:
-#             buildDependsOn: ${{ component.unifiedBuild.buildDependsOn }}
+            buildDependsOn: ${{ component.unifiedBuild.buildDependsOn }}
-#             downstreamAggregateNames: ${{ parameters.downstreamAggregateNames }}+${{ component.unifiedBuild.downstreamAggregateNames }}
+            downstreamAggregateNames: ${{ parameters.downstreamAggregateNames }}+${{ component.unifiedBuild.downstreamAggregateNames }}
-#           ${{ else }}:
+          ${{ else }}:
-#             buildDependsOn: ${{ component.buildDependsOn }}
+            buildDependsOn: ${{ component.buildDependsOn }}
-#             downstreamAggregateNames: ${{ parameters.downstreamAggregateNames }}+${{ parameters.componentName }}
+            downstreamAggregateNames: ${{ parameters.downstreamAggregateNames }}+${{ parameters.componentName }}

.azuredevops/components/rocFFT.yml

@@ -166,6 +166,7 @@ jobs:
     workspace:
       clean: all
     steps:
+    - checkout: none
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
       parameters:
         aptPackages: ${{ parameters.aptPackages }}

.azuredevops/components/rocMLIR.yml

@@ -27,6 +27,7 @@ parameters:
     - numpy
     - tomli
     - scipy
+    - pybind11
 - name: rocmDependencies
   type: object
   default:

.azuredevops/components/rocPRIM.yml

@@ -210,7 +210,7 @@ jobs:
         parameters:
           componentName: ${{ parameters.componentName }}
           testDir: '$(Agent.BuildDirectory)/rocm/bin/rocprim'
-          extraTestParameters: '-I ${{ job.shard }},,${{ job.shardCount }}'
+          extraTestParameters: '-I ${{ job.shard }},,${{ job.shardCount }} -E device_merge_inplace'
           os: ${{ job.os }}
       - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/docker-container.yml
         parameters:

.azuredevops/components/rocPyDecode.yml

@@ -36,6 +36,7 @@ parameters:
     - clr
     - llvm-project
     - rocDecode
+    - rocJPEG
     - rocm-cmake
     - rocm-core
     - rocminfo
@@ -192,9 +193,9 @@ jobs:
       inputs:
         itemPattern: '**/*.whl'
         targetPath: $(Agent.BuildDirectory)
-    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/checkout.yml
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/local-artifact-download.yml
       parameters:
-        checkoutRepo: ${{ parameters.checkoutRepo }}
+        gpuTarget: ${{ job.target }}
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-aqlprofile.yml
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
       parameters:
@@ -221,25 +222,17 @@ jobs:
     - task: CMake@1
       displayName: 'rocPyDecode Test CMake Flags'
       inputs:
+        workingDirectory: $(Agent.BuildDirectory)/rocm/share/rocpydecode/tests
         cmakeArgs: >-
           -DCMAKE_PREFIX_PATH=$(Agent.BuildDirectory)/rocm;$(PYTHON_USER_SITE)/pybind11;$(PYTHON_DIST_PACKAGES)/pybind11;$(PYBIND11_PATH)
           -DCMAKE_BUILD_TYPE=Release
           -DGPU_TARGETS=${{ job.target }}
-          ..
+          .
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/gpu-diagnostics.yml
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/test.yml
       parameters:
         componentName: rocPyDecode
-        testDir: $(Build.SourcesDirectory)/build
+        testDir: $(Agent.BuildDirectory)/rocm/share/rocpydecode/tests
-  # sudo required for pip install but screws up permissions for next pipeline run
-    - task: Bash@3
-      displayName: Clean up test environment
-      condition: always()
-      inputs:
-        targetType: inline
-        script: |
-          pip uninstall -y rocPyDecode
-          pip uninstall -y hip-python
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/docker-container.yml
       parameters:
         aptPackages: ${{ parameters.aptPackages }}

.azuredevops/components/rocSOLVER.yml

@@ -33,13 +33,11 @@ parameters:
   type: object
   default:
     - cmake
-    - ninja-build
-    - libsuitesparse-dev
     - gfortran
-    - libfmt-dev
     - git
-    - googletest
+    - libfmt-dev
-    - libgtest-dev
+    - libsuitesparse-dev
+    - ninja-build
     - python3-pip
 - name: rocmDependencies
   type: object
@@ -75,13 +73,13 @@ parameters:
       - { os: ubuntu2204, packageManager: apt, target: gfx942 }
       - { os: ubuntu2204, packageManager: apt, target: gfx90a }
       - { os: ubuntu2204, packageManager: apt, target: gfx1201 }
-      # - { os: ubuntu2204, packageManager: apt, target: gfx1100 }
+      - { os: ubuntu2204, packageManager: apt, target: gfx1100 }
       - { os: ubuntu2204, packageManager: apt, target: gfx1030 }
-      # - { os: almalinux8, packageManager: dnf, target: gfx942 }
+      - { os: almalinux8, packageManager: dnf, target: gfx942 }
-      # - { os: almalinux8, packageManager: dnf, target: gfx90a }
+      - { os: almalinux8, packageManager: dnf, target: gfx90a }
-      # - { os: almalinux8, packageManager: dnf, target: gfx1201 }
+      - { os: almalinux8, packageManager: dnf, target: gfx1201 }
-      # - { os: almalinux8, packageManager: dnf, target: gfx1100 }
+      - { os: almalinux8, packageManager: dnf, target: gfx1100 }
-      # - { os: almalinux8, packageManager: dnf, target: gfx1030 }
+      - { os: almalinux8, packageManager: dnf, target: gfx1030 }
     testJobs:
       - { os: ubuntu2204, packageManager: apt, target: gfx942 }
       - { os: ubuntu2204, packageManager: apt, target: gfx90a }
@@ -119,6 +117,10 @@ jobs:
         targetType: inline
         script: git clone --depth 1 --branch v3.9.1 https://github.com/Reference-LAPACK/lapack
         workingDirectory: '$(Build.SourcesDirectory)'
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-vendor.yml
+      parameters:
+        dependencyList:
+          - gtest
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
       parameters:
         checkoutRef: ${{ parameters.checkoutRef }}
@@ -134,6 +136,7 @@ jobs:
         os: ${{ job.os }}
         extraBuildFlags: >-
           -DCMAKE_BUILD_TYPE=Release
+          -DCMAKE_POSITION_INDEPENDENT_CODE=ON
           -DCMAKE_Fortran_FLAGS=-fno-optimize-sibling-calls
           -DBUILD_TESTING=OFF
           -DCBLAS=ON
@@ -146,7 +149,7 @@ jobs:
       parameters:
         os: ${{ job.os }}
         extraBuildFlags: >-
-          -DCMAKE_PREFIX_PATH=$(Agent.BuildDirectory)/rocm;$(Pipeline.Workspace)/deps-install
+          -DCMAKE_PREFIX_PATH=$(Agent.BuildDirectory)/rocm;$(Pipeline.Workspace)/deps-install;$(Agent.BuildDirectory)/vendor
           -DCMAKE_CXX_COMPILER=$(Agent.BuildDirectory)/rocm/llvm/bin/amdclang++
           -DCMAKE_C_COMPILER=$(Agent.BuildDirectory)/rocm/llvm/bin/amdclang
           -DAMDGPU_TARGETS=${{ job.target }}
@@ -191,6 +194,7 @@ jobs:
       workspace:
         clean: all
       steps:
+      - checkout: none
       - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
         parameters:
           aptPackages: ${{ parameters.aptPackages }}

.azuredevops/components/rocSPARSE.yml

@@ -1,10 +1,29 @@
 parameters:
+- name: componentName
+  type: string
+  default: rocSPARSE
 - name: checkoutRepo
   type: string
   default: 'self'
 - name: checkoutRef
   type: string
   default: ''
+# monorepo related parameters
+- name: sparseCheckoutDir
+  type: string
+  default: ''
+- name: triggerDownstreamJobs
+  type: boolean
+  default: false
+- name: downstreamAggregateNames
+  type: string
+  default: ''
+- name: buildDependsOn
+  type: object
+  default: null
+- name: unifiedBuild
+  type: boolean
+  default: false
 # set to true if doing full build of ROCm stack
 # and dependencies are pulled from same pipeline
 - name: aggregatePipeline
@@ -13,27 +32,25 @@ parameters:
 - name: aptPackages
   type: object
   default:
-    - python3-pip
     - cmake
-    - ninja-build
-    - libboost-program-options-dev
-    - googletest
-    - libfftw3-dev
-    - git
     - gfortran
-    - libgtest-dev
+    - git
+    - libboost-program-options-dev
     - libdrm-dev
+    - libfftw3-dev
+    - ninja-build
+    - python3-pip
 - name: rocmDependencies
   type: object
   default:
-    - rocm-cmake
-    - llvm-project
-    - ROCR-Runtime
     - clr
+    - llvm-project
     - rocBLAS
+    - rocm-cmake
     - rocminfo
     - rocPRIM
     - rocprofiler-register
+    - ROCR-Runtime
     - roctracer
 - name: rocmTestDependencies
   type: object
@@ -52,19 +69,39 @@ parameters:
   type: object
   default:
     buildJobs:
-      - gfx942:
+      - { os: ubuntu2204, packageManager: apt, target: gfx942 }
-        target: gfx942
+      - { os: ubuntu2204, packageManager: apt, target: gfx90a }
-      - gfx90a:
+      - { os: ubuntu2204, packageManager: apt, target: gfx1201 }
-        target: gfx90a
+      - { os: ubuntu2204, packageManager: apt, target: gfx1100 }
+      - { os: ubuntu2204, packageManager: apt, target: gfx1030 }
     testJobs:
-      - gfx942:
+      - { os: ubuntu2204, packageManager: apt, target: gfx942 }
-        target: gfx942
+      - { os: ubuntu2204, packageManager: apt, target: gfx90a }
-      - gfx90a:
+- name: downstreamComponentMatrix
-        target: gfx90a
+  type: object
+  default:
+    - hipSPARSE:
+      name: hipSPARSE
+      sparseCheckoutDir: projects/hipsparse
+      skipUnifiedBuild: 'false'
+      buildDependsOn:
+        - rocSPARSE_build
+    # hipSOLVER depends on both rocSOLVER and rocSPARSE
+    # for a unified build, rocSOLVER will be the one to call hipSOLVER
+    # - hipSOLVER:
+    #   name: hipSOLVER
+    #   sparseCheckoutDir: projects/hipsolver
+    #   skipUnifiedBuild: 'true'
+    #   buildDependsOn:
+    #     - rocSPARSE_build
 
 jobs:
 - ${{ each job in parameters.jobMatrix.buildJobs }}:
-  - job: rocSPARSE_build_${{ job.target }}
+  - job: ${{ parameters.componentName }}_build_${{ job.os }}_${{ job.target }}
+    ${{ if parameters.buildDependsOn }}:
+      dependsOn:
+        - ${{ each build in parameters.buildDependsOn }}:
+          - ${{ build }}_${{ job.os }}_${{ job.target }}
     variables:
     - group: common
     - template: /.azuredevops/variables-global.yml
@@ -77,22 +114,32 @@ jobs:
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
       parameters:
         aptPackages: ${{ parameters.aptPackages }}
+        packageManager: ${{ job.packageManager }}
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/preamble.yml
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/checkout.yml
       parameters:
         checkoutRepo: ${{ parameters.checkoutRepo }}
+        sparseCheckoutDir: ${{ parameters.sparseCheckoutDir }}
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-vendor.yml
+      parameters:
+        dependencyList:
+          - gtest
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
       parameters:
         checkoutRef: ${{ parameters.checkoutRef }}
         dependencyList: ${{ parameters.rocmDependencies }}
         gpuTarget: ${{ job.target }}
+        os: ${{ job.os }}
         aggregatePipeline: ${{ parameters.aggregatePipeline }}
+        ${{ if parameters.triggerDownstreamJobs }}:
+          downstreamAggregateNames: ${{ parameters.downstreamAggregateNames }}
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
       parameters:
+        os: ${{ job.os }}
         extraBuildFlags: >-
-          -DCMAKE_CXX_COMPILER=$(Agent.BuildDirectory)/rocm/bin/hipcc
+          -DCMAKE_PREFIX_PATH=$(Agent.BuildDirectory)/rocm;$(Agent.BuildDirectory)/vendor
-          -DCMAKE_C_COMPILER=$(Agent.BuildDirectory)/rocm/bin/hipcc
+          -DCMAKE_CXX_COMPILER=$(Agent.BuildDirectory)/rocm/bin/amdclang++
-          -DCMAKE_PREFIX_PATH=$(Agent.BuildDirectory)/rocm
+          -DCMAKE_C_COMPILER=$(Agent.BuildDirectory)/rocm/bin/amdclang
           -DROCM_PATH=$(Agent.BuildDirectory)/rocm
           -DCMAKE_BUILD_TYPE=Release
           -DAMDGPU_TARGETS=${{ job.target }}
@@ -103,68 +150,94 @@ jobs:
           -GNinja
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/manifest.yml
       parameters:
-        artifactName: rocSPARSE
+        componentName: ${{ parameters.componentName }}
+        sparseCheckoutDir: ${{ parameters.sparseCheckoutDir }}
         gpuTarget: ${{ job.target }}
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/artifact-upload.yml
       parameters:
-        artifactName: rocSPARSE
+        componentName: ${{ parameters.componentName }}
         gpuTarget: ${{ job.target }}
+        os: ${{ job.os }}
         publish: false
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/artifact-prepare-package.yml
       parameters:
-        sourceDir: $(Build.SourcesDirectory)/build/clients
+        sourceDir: $(Agent.BuildDirectory)/s/build/clients
         contentsString: matrices/**
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/artifact-upload.yml
       parameters:
+        componentName: ${{ parameters.componentName }}
         artifactName: testMatrices
         gpuTarget: ${{ job.target }}
+        os: ${{ job.os }}
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/artifact-links.yml
-    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/docker-container.yml
+    - ${{ if eq(job.os, 'ubuntu2204') }}:
-      parameters:
+      - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/docker-container.yml
-        aptPackages: ${{ parameters.aptPackages }}
+        parameters:
-        gpuTarget: ${{ job.target }}
+          aptPackages: ${{ parameters.aptPackages }}
-        extraEnvVars:
+          gpuTarget: ${{ job.target }}
-          - HIP_ROCCLR_HOME:::/home/user/workspace/rocm
+          extraEnvVars:
+            - HIP_ROCCLR_HOME:::/home/user/workspace/rocm
 
-- ${{ each job in parameters.jobMatrix.testJobs }}:
+- ${{ if eq(parameters.unifiedBuild, False) }}:
-  - job: rocSPARSE_test_${{ job.target }}
+  - ${{ each job in parameters.jobMatrix.testJobs }}:
-    timeoutInMinutes: 90
+    - job: ${{ parameters.componentName }}_test_${{ job.os }}_${{ job.target }}
-    dependsOn: rocSPARSE_build_${{ job.target }}
+      timeoutInMinutes: 120
-    condition:
+      dependsOn: ${{ parameters.componentName }}_build_${{ job.os }}_${{ job.target }}
-      and(succeeded(),
+      condition:
-        eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
+        and(succeeded(),
-        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName'])),
+          eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
-        eq(${{ parameters.aggregatePipeline }}, False)
+          not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), '${{ parameters.componentName }}')),
-      )
+          eq(${{ parameters.aggregatePipeline }}, False)
-    variables:
+        )
-    - group: common
+      variables:
-    - template: /.azuredevops/variables-global.yml
+      - group: common
-    pool: ${{ job.target }}_test_pool
+      - template: /.azuredevops/variables-global.yml
-    workspace:
+      pool: ${{ job.target }}_test_pool
-      clean: all
+      workspace:
-    steps:
+        clean: all
-    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
+      steps:
-      parameters:
+      - checkout: none
-        aptPackages: ${{ parameters.aptPackages }}
+      - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
-    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/preamble.yml
+        parameters:
-    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/local-artifact-download.yml
+          aptPackages: ${{ parameters.aptPackages }}
-      parameters:
+          packageManager: ${{ job.packageManager }}
-        gpuTarget: ${{ job.target }}
+      - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/preamble.yml
-    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-aqlprofile.yml
+      - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/local-artifact-download.yml
-    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+        parameters:
-      parameters:
+          preTargetFilter: ${{ parameters.componentName }}
-        checkoutRef: ${{ parameters.checkoutRef }}
+          gpuTarget: ${{ job.target }}
-        dependencyList: ${{ parameters.rocmTestDependencies }}
+          os: ${{ job.os }}
-        gpuTarget: ${{ job.target }}
+      - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-aqlprofile.yml
-    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/gpu-diagnostics.yml
+      - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
-    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/test.yml
+        parameters:
-      parameters:
+          checkoutRef: ${{ parameters.checkoutRef }}
-        componentName: rocSPARSE
+          dependencyList: ${{ parameters.rocmTestDependencies }}
-        testDir: '$(Agent.BuildDirectory)/rocm/bin'
+          gpuTarget: ${{ job.target }}
-        testExecutable: './rocsparse-test'
+          os: ${{ job.os }}
-        testParameters: '--gtest_filter="*quick*" --gtest_output=xml:./test_output.xml --gtest_color=yes'
+          ${{ if parameters.triggerDownstreamJobs }}:
-    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/docker-container.yml
+            downstreamAggregateNames: ${{ parameters.downstreamAggregateNames }}
-      parameters:
+      - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/gpu-diagnostics.yml
-        aptPackages: ${{ parameters.aptPackages }}
+      - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/test.yml
-        environment: test
+        parameters:
-        gpuTarget: ${{ job.target }}
+          componentName: ${{ parameters.componentName }}
+          os: ${{ job.os }}
+          testDir: '$(Agent.BuildDirectory)/rocm/bin'
+          testExecutable: './rocsparse-test'
+          testParameters: '--gtest_filter="*quick*" --gtest_output=xml:./test_output.xml --gtest_color=yes'
+      - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/docker-container.yml
+        parameters:
+          aptPackages: ${{ parameters.aptPackages }}
+          environment: test
+          gpuTarget: ${{ job.target }}
+
+- ${{ if parameters.triggerDownstreamJobs }}:
+  - ${{ each component in parameters.downstreamComponentMatrix }}:
+    - ${{ if not(and(parameters.unifiedBuild, eq(component.skipUnifiedBuild, 'true'))) }}:
+      - template: /.azuredevops/components/${{ component.name }}.yml@pipelines_repo
+        parameters:
+          checkoutRepo: ${{ parameters.checkoutRepo }}
+          sparseCheckoutDir: ${{ component.sparseCheckoutDir }}
+          buildDependsOn: ${{ component.buildDependsOn }}
+          downstreamAggregateNames: ${{ parameters.downstreamAggregateNames }}+${{ parameters.componentName }}
+          triggerDownstreamJobs: true
+          unifiedBuild: ${{ parameters.unifiedBuild }}

.azuredevops/components/rocm-examples.yml

@@ -184,7 +184,7 @@ jobs:
       parameters:
         componentName: rocm-examples
         testDir: $(Build.SourcesDirectory)/build
-        testParameters: '--output-on-failure --force-new-ctest-process --output-junit test_output.xml --exclude-regex "rocfft_callback"'
+        testParameters: '--output-on-failure --force-new-ctest-process --output-junit test_output.xml'
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/docker-container.yml
       parameters:
         aptPackages: ${{ parameters.aptPackages }}

.azuredevops/components/rocminfo.yml

@@ -67,7 +67,6 @@ jobs:
         checkoutRef: ${{ parameters.checkoutRef }}
         dependencyList: ${{ parameters.rocmDependencies }}
         aggregatePipeline: ${{ parameters.aggregatePipeline }}
-        skipLlvmSymlink: true
         os: ${{ job.os }}
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
       parameters:

.azuredevops/nightly/pytorch.yml

@@ -407,7 +407,6 @@ jobs:
       dependencyList: ${{ parameters.rocmTestDependencies }}
       gpuTarget: $(JOB_GPU_TARGET)
       dependencySource: staging
-      skipLlvmSymlink: true
 # get sources to run test scripts
   - task: Bash@3
     displayName: git clone upstream pytorch

.azuredevops/nightly/rocm-nightly.yml

@@ -119,7 +119,6 @@ jobs:
         dependencyList: ${{ parameters.rocmDependencies }}
         os: ${{ job.os }}
         gpuTarget: ${{ job.target }}
-        skipLibraryLinking: true
     - script: df -h
       displayName: System disk space after ROCm
     - script: du -sh $(Agent.BuildDirectory)/rocm

.azuredevops/templates/steps/artifact-download.yml

@@ -12,6 +12,9 @@ parameters:
 - name: fileFilter
   type: string
   default: ''
+- name: extractAndDeleteFiles
+  type: boolean
+  default: true
 # set to true if doing full build of ROCm stack
 # and dependencies are pulled from same pipeline
 - name: aggregatePipeline
@@ -22,34 +25,32 @@ steps:
 - task: DownloadPipelineArtifact@2
   displayName: Download ${{ parameters.componentName }}
   inputs:
+    itemPattern: '**/*${{ parameters.componentName }}*${{ parameters.fileFilter }}*'
+    targetPath: '$(Pipeline.Workspace)/d'
+    allowPartiallySucceededBuilds: true
     ${{ if parameters.aggregatePipeline }}:
       buildType: 'current'
-      itemPattern: '**/${{ parameters.componentName }}*${{ parameters.fileFilter }}*'
-      allowPartiallySucceededBuilds: true
-      targetPath: '$(Pipeline.Workspace)/d'
     ${{ else }}:
       buildType: 'specific'
       project: ROCm-CI
       specificBuildWithTriggering: true
-      allowPartiallySucceededBuilds: true
       definition: ${{ parameters.pipelineId }}
-      itemPattern: '**/*${{ parameters.fileFilter }}*'
-      targetPath: '$(Pipeline.Workspace)/d'
       branchName: refs/heads/${{ parameters.branchName }}
       ${{ if eq(parameters.componentName, 'aomp') }}:
         buildVersionToDownload: latest # aomp trigger lives in ROCm/ROCm, so cannot use ROCm/aomp branch names
       ${{ else }}:
         buildVersionToDownload: latestFromBranch
-- task: ExtractFiles@1
+- ${{ if eq(parameters.extractAndDeleteFiles, true) }}:
-  displayName: Extract ${{ parameters.componentName }}
+  - task: ExtractFiles@1
-  inputs:
+    displayName: Extract ${{ parameters.componentName }}
-    archiveFilePatterns: '$(Pipeline.Workspace)/d/**/*.tar.gz'
+    inputs:
-    destinationFolder: '$(Agent.BuildDirectory)/rocm'
+      archiveFilePatterns: '$(Pipeline.Workspace)/d/**/*.tar.gz'
-    cleanDestinationFolder: false
+      destinationFolder: '$(Agent.BuildDirectory)/rocm'
-    overwriteExistingFiles: true
+      cleanDestinationFolder: false
-- task: DeleteFiles@1
+      overwriteExistingFiles: true
-  displayName: Cleanup Compressed ${{ parameters.componentName }}
+  - task: DeleteFiles@1
-  inputs:
+    displayName: Clean up Compressed ${{ parameters.componentName }}
-    SourceFolder: '$(Pipeline.Workspace)/d'
+    inputs:
-    Contents: '**/*.tar.gz'
+      SourceFolder: '$(Pipeline.Workspace)/d'
-    RemoveDotFiles: true
+      Contents: '**/*.tar.gz'
+      RemoveDotFiles: true

.azuredevops/templates/steps/artifact-links.yml

@@ -15,8 +15,8 @@ steps:
       URL_BEGIN="https://artprodcus3.artifacts.visualstudio.com/"
       URL_MIDDLE="/_apis/artifact/"
       URL_END="/content?format=file&subPath=%2F"
-      FORMATTED_JOB_NAME=$(echo $(Agent.JobName) | sed 's/ /./g; s/[-_]//g')
+      ARTIFACT_NAME="$(Agent.JobName)_$(System.JobAttempt)"
-      ARTIFACT_STRING="pipelineartifact://ROCm-CI/projectId/$(DOWNLOAD_PROJECT_ID)/buildId/$(Build.BuildId)/artifactName/${FORMATTED_JOB_NAME}"
+      ARTIFACT_STRING="pipelineartifact://ROCm-CI/projectId/$(DOWNLOAD_PROJECT_ID)/buildId/$(Build.BuildId)/artifactName/${ARTIFACT_NAME}"
       ENCODED_STRING=$(echo -n "${ARTIFACT_STRING}" | base64 -w 0)
       PADDING_COUNT=$(echo -n "${ENCODED_STRING}" | awk -F= '{print NF-1}')
       if [ "$PADDING_COUNT" -gt 0 ]; then

.azuredevops/templates/steps/artifact-upload.yml

@@ -46,5 +46,6 @@ steps:
     displayName: '${{ parameters.artifactName }} Publish'
     retryCountOnTaskFailure: 3
     inputs:
-      artifactName: ${{ parameters.componentName }}_$(Build.BuildId)_$(Build.BuildNumber)_${{ parameters.os }}_${{ parameters.gpuTarget }}_${{ parameters.artifactName }}_$(System.JobAttempt)
+      # if this artifact name is changed, please also update $ARTIFACT_URL inside miopen-get-ck-build.yml
+      artifactName: $(Agent.JobName)_$(System.JobAttempt)
       targetPath: '$(Build.ArtifactStagingDirectory)'

.azuredevops/templates/steps/dependencies-aocl.yml

@@ -1,10 +1,15 @@
 parameters:
+- name: os
+  type: string
+  default: ubuntu2204
 - name: repositoryUrl
   type: string
   default: https://download.amd.com/developer/eula/aocl/aocl-4-2
 - name: packageName
-  type: string
+  type: object
-  default: aocl-linux-gcc-4.2.0_1_amd64.deb
+  default:
+    ubuntu2204: aocl-linux-gcc-4.2.0_1_amd64.deb
+    almalinux8: aocl-linux-gcc-4.2.0-1.x86_64.rpm
 
 steps:
 - task: Bash@3
@@ -12,16 +17,19 @@ steps:
   inputs:
     targetType: inline
     workingDirectory: $(Pipeline.Workspace)
-    script: wget -nv ${{ parameters.repositoryUrl }}/${{ parameters.packageName }}
+    script: wget -nv ${{ parameters.repositoryUrl }}/${{ parameters.packageName[parameters.os] }}
 - task: Bash@3
   displayName: Install AOCL
   inputs:
     targetType: inline
     workingDirectory: $(Pipeline.Workspace)
-    script: sudo apt install -y ./${{ parameters.packageName }}
+    ${{ if eq(parameters.os, 'ubuntu2204') }}:
+      script: sudo apt install -y ./${{ parameters.packageName[parameters.os] }}
+    ${{ elseif eq(parameters.os, 'almalinux8') }}:
+      script: sudo dnf install -y ./${{ parameters.packageName[parameters.os] }}
 - task: Bash@3
   displayName: Clean up AOCL
   inputs:
     targetType: inline
     workingDirectory: $(Pipeline.Workspace)
-    script: rm -f ${{ parameters.packageName }}
+    script: rm -f ${{ parameters.packageName[parameters.os] }}

.azuredevops/templates/steps/dependencies-dnf.yml

@@ -52,6 +52,7 @@ parameters:
     libexpat-dev: expat-devel
     libffi-dev: libffi-devel
     libfftw3-dev: fftw-devel
+    libfmt-dev: fmt-devel
     libgmp-dev: gmp-devel
     liblzma-dev: xz-devel
     libmpfr-dev: mpfr-devel

.azuredevops/templates/steps/dependencies-rocm.yml

@@ -19,16 +19,6 @@ parameters:
 - name: gpuTarget
   type: string
   default: ''
-# set to true if you're calling this template file multiple files in same pipeline
-# only leave last call false to optimize sequence
-- name: skipLibraryLinking
-  type: boolean
-  default: false
-# set to true if llvm-project is not downloaded in a particular call
-# or if you just don't want the symlink
-- name: skipLlvmSymlink
-  type: boolean
-  default: false
 # set to true if dlopen calls for HIP libraries are causing failures
 # because they do not follow shared library symlink convention
 - name: setupHIPLibrarySymlinks
@@ -367,6 +357,7 @@ steps:
         componentName: ${{ split(dependency, ':')[0] }}
         pipelineId: ${{ parameters.componentVarList[split(dependency, ':')[0]].pipelineId }}
         aggregatePipeline: ${{ parameters.aggregatePipeline }}
+        extractAndDeleteFiles: false
         ${{ if parameters.componentVarList[split(dependency, ':')[0]].hasGpuTarget }}:
           fileFilter: "${{ split(dependency, ':')[1] }}*_${{ parameters.os }}_${{ parameters.gpuTarget }}"
         # dependencySource = staging
@@ -405,6 +396,7 @@ steps:
         componentName: ${{ dependency }}
         pipelineId: ${{ parameters.componentVarList[dependency].pipelineId }}
         aggregatePipeline: ${{ parameters.aggregatePipeline }}
+        extractAndDeleteFiles: false
         ${{ if parameters.componentVarList[dependency].hasGpuTarget }}:
           fileFilter: ${{ parameters.os }}_${{ parameters.gpuTarget }}
         ${{ else }}:
@@ -430,8 +422,20 @@ steps:
         # default = staging
         ${{ else }}:
           branchName: ${{ parameters.componentVarList[dependency].stagingBranch }}
-# Set link to redirect llvm folder
+- task: ExtractFiles@1
-- ${{ if eq(parameters.skipLlvmSymlink, false) }}:
+  displayName: Extract ROCm artifacts
+  inputs:
+    archiveFilePatterns: $(Pipeline.Workspace)/d/**/*.tar.gz
+    destinationFolder: $(Agent.BuildDirectory)/rocm
+    cleanDestinationFolder: false
+    overwriteExistingFiles: true
+- task: DeleteFiles@1
+  displayName: Clean up ROCm artifacts
+  inputs:
+    SourceFolder: $(Pipeline.Workspace)/d
+    Contents: '**/*.tar.gz'
+    RemoveDotFiles: true
+- ${{ if containsValue(parameters.dependencyList, 'llvm-project') }}:
   - task: Bash@3
     displayName: Symlink from rocm/llvm to rocm/lib/llvm
     inputs:
@@ -439,6 +443,7 @@ steps:
       script: |
         sudo mkdir -p $(Agent.BuildDirectory)/rocm/lib
         sudo ln -sr $(Agent.BuildDirectory)/rocm/llvm $(Agent.BuildDirectory)/rocm/lib/llvm
+        echo "Created symlink from rocm/llvm to rocm/lib/llvm"
   - task: Bash@3
     displayName: Symlink executables from rocm/llvm/bin to rocm/bin
     inputs:
@@ -446,7 +451,14 @@ steps:
       script: |
         for file in amdclang amdclang++ amdclang-cl amdclang-cpp amdflang amdlld aompcc mygpu mycpu offload-arch; do
           sudo ln -sr $(Agent.BuildDirectory)/rocm/llvm/bin/$file $(Agent.BuildDirectory)/rocm/bin/$file
+          echo "Created symlink from rocm/llvm/bin/$file to rocm/bin/$file"
         done
+- ${{ if containsValue(parameters.dependencyList, 'rocm-core') }}:
+  - task: Bash@3
+    displayName: Print rocm/.info/version
+    inputs:
+      targetType: inline
+      script: cat $(Agent.BuildDirectory)/rocm/.info/version
 # dlopen calls within a ctest or pytest sequence runs into issues when shared library symlink convention is not followed
 # the convention is as follows:
 # unversioned .so is a symlink to major version .so
@@ -483,17 +495,16 @@ steps:
   inputs:
     targetType: inline
     script: ls -la1R $(Agent.BuildDirectory)/rocm
-- ${{ if eq(parameters.skipLibraryLinking, false) }}:
+- task: Bash@3
-  - task: Bash@3
+  displayName: 'Link ROCm shared libraries'
-    displayName: 'Link ROCm shared libraries'
+  inputs:
-    inputs:
+    targetType: inline
-      targetType: inline
+    # OS ignores if the ROCm lib folder shows up more than once
-# OS ignores if the ROCm lib folder shows up more than once
+    script: |
-      script: |
+      echo $(Agent.BuildDirectory)/rocm/lib | sudo tee -a /etc/ld.so.conf.d/rocm-ci.conf
-        echo $(Agent.BuildDirectory)/rocm/lib | sudo tee -a /etc/ld.so.conf.d/rocm-ci.conf
+      echo $(Agent.BuildDirectory)/rocm/llvm/lib | sudo tee -a /etc/ld.so.conf.d/rocm-ci.conf
-        echo $(Agent.BuildDirectory)/rocm/llvm/lib | sudo tee -a /etc/ld.so.conf.d/rocm-ci.conf
+      echo $(Agent.BuildDirectory)/rocm/lib64 | sudo tee -a /etc/ld.so.conf.d/rocm-ci.conf
-        echo $(Agent.BuildDirectory)/rocm/lib64 | sudo tee -a /etc/ld.so.conf.d/rocm-ci.conf
+      echo $(Agent.BuildDirectory)/rocm/llvm/lib64 | sudo tee -a /etc/ld.so.conf.d/rocm-ci.conf
-        echo $(Agent.BuildDirectory)/rocm/llvm/lib64 | sudo tee -a /etc/ld.so.conf.d/rocm-ci.conf
+      sudo cat /etc/ld.so.conf.d/rocm-ci.conf
-        sudo cat /etc/ld.so.conf.d/rocm-ci.conf
+      sudo ldconfig -v
-        sudo ldconfig -v
+      ldconfig -p
-        ldconfig -p

.azuredevops/templates/steps/manifest.yml

@@ -23,13 +23,14 @@ steps:
   inputs:
     targetType: inline
     script: |
-      sudo apt-get install -y jq
+      ${{ iif(or(eq(parameters.os, 'ubuntu2204'), eq(parameters.os, 'ubuntu2404')), 'sudo apt-get install -y jq', '') }}
 
       # RESOURCES_REPOSITORIES is a runtime variable (not an env var!) that contains quotations and newlines
       # So we need to save it to a file to properly preserve its formatting and contents
       cat <<EOF > resources.repositories
       $(RESOURCES_REPOSITORIES)
       EOF
+      echo "Value of resources.repositories:"
       cat resources.repositories
 
       IS_TAG_BUILD=$(jq 'has("release_repo")' resources.repositories)
@@ -66,8 +67,6 @@ steps:
         )
       ' resources.repositories)
 
-      manifest_json=$(Build.ArtifactStagingDirectory)/manifest_${{ parameters.componentName }}_$(Build.BuildId)_$(Build.BuildNumber)_${{ parameters.os }}_${{ parameters.gpuTarget }}_${{ parameters.artifactName }}.json
-
       dependencies=()
       for manifest_file in $(Pipeline.Workspace)/d/**/manifest_*.json; do
         echo "Processing $manifest_file"
@@ -78,6 +77,10 @@ steps:
       done
       dependencies_json=$(printf '%s\n' "${dependencies[@]}" | jq -s '.')
 
+      manifest_filename="manifest_${{ parameters.componentName }}_$(Build.BuildId)_$(Build.BuildNumber)_${{ parameters.os }}_${{ parameters.gpuTarget }}_${{ parameters.artifactName }}"
+      echo "##vso[task.setvariable variable=manifest_filename]$manifest_filename"
+      manifest_json=$(Build.ArtifactStagingDirectory)/$manifest_filename.json
+
       jq -n \
         --argjson current "$current" \
         --argjson dependencies "$dependencies_json" \
@@ -111,8 +114,14 @@ steps:
         ')
       dependencies_rows=$(echo $dependencies_rows)
       echo "##vso[task.setvariable variable=dependencies_rows;]$dependencies_rows"
-
+- task: Bash@3
-      cat $manifest_json
+  displayName: Print manifest.json
+  condition: always()
+  continueOnError: true
+  inputs:
+    targetType: inline
+    script: |
+      cat $(Build.ArtifactStagingDirectory)/$(manifest_filename).json
 - task: Bash@3
   displayName: Create manifest.html
   condition: always()
@@ -120,10 +129,10 @@ steps:
   inputs:
     targetType: inline
     script: |
-      manifest_html=$(Build.ArtifactStagingDirectory)/manifest_${{ parameters.componentName }}_$(Build.BuildId)_$(Build.BuildNumber)_${{ parameters.os }}_${{ parameters.gpuTarget }}_${{ parameters.artifactName }}.html
+      manifest_html="$(Build.ArtifactStagingDirectory)/$(manifest_filename).html"
       cat <<EOF > $manifest_html
       <html>
-      <h1>Manifest</h1>
+      <h1>$(manifest_filename)</h1>
       <h2>Current</h2>
       <table border="1">
       <tr>
@@ -163,7 +172,7 @@ steps:
   continueOnError: true
   inputs:
     tabName: Manifest
-    reportDir: $(Build.ArtifactStagingDirectory)/manifest_${{ parameters.componentName }}_$(Build.BuildId)_$(Build.BuildNumber)_${{ parameters.os }}_${{ parameters.gpuTarget }}_${{ parameters.artifactName }}.html
+    reportDir: $(Build.ArtifactStagingDirectory)/$(manifest_filename).html
 - task: Bash@3
   displayName: Save manifest artifact file name
   condition: always()
@@ -172,5 +181,5 @@ steps:
     workingDirectory: $(Pipeline.Workspace)
     targetType: inline
     script: |
-      echo "manifest_${{ parameters.componentName }}_$(Build.BuildId)_$(Build.BuildNumber)_${{ parameters.os }}_${{ parameters.gpuTarget }}_${{ parameters.artifactName }}.html" >> pipelineArtifacts.txt
+      echo "$(manifest_filename).html" >> pipelineArtifacts.txt
-      echo "manifest_${{ parameters.componentName }}_$(Build.BuildId)_$(Build.BuildNumber)_${{ parameters.os }}_${{ parameters.gpuTarget }}_${{ parameters.artifactName }}.json" >> pipelineArtifacts.txt
+      echo "$(manifest_filename).json" >> pipelineArtifacts.txt

.azuredevops/templates/steps/miopen-get-ck-build.yml

@@ -17,7 +17,6 @@ steps:
     script: |
       AZ_API="https://dev.azure.com/ROCm-CI/ROCm-CI/_apis"
       GH_API="https://api.github.com/repos/ROCm"
-      ARTIFACT_NAME="composablekernelbuild${{ parameters.gpuTarget }}"
       EXIT_CODE=0
 
       # Try to find an Azure build for the specific CK commit called out in MIOpen's requirements.txt
@@ -39,8 +38,15 @@ steps:
         echo "Found specific CK build ID: $CK_BUILD_ID"
       fi
 
-      AZURE_URL="$AZ_API/build/builds/$CK_BUILD_ID/artifacts?artifactName=$ARTIFACT_NAME&api-version=7.1"
+      AZURE_URL="$AZ_API/build/builds/$CK_BUILD_ID/artifacts?api-version=7.1"
-      ARTIFACT_URL=$(curl -s $AZURE_URL | jq '.resource.downloadUrl' | tr -d '"')
+      ARTIFACT_URL=$(curl -s $AZURE_URL | \
+        jq --arg gfx "${{ parameters.gpuTarget }}" '
+          .value
+          | map(select(.name | test($gfx)))
+          | max_by(.name | capture("_(?<dropNumber>\\d+)").dropNumber | tonumber)
+          | .resource.downloadUrl
+        ' | \
+        tr -d '"')
 
       # If using the specific CK commit and it doesn't have any valid artifacts, use latest successful CK build instead
       if { [[ -z "$ARTIFACT_URL" ]] || [[ "$ARTIFACT_URL" == "null" ]]; } && [[ $EXIT_CODE -eq 0 ]]; then
@@ -48,8 +54,15 @@ steps:
         LATEST_BUILD_URL="$AZ_API/build/builds?definitions=$(COMPOSABLE_KERNEL_PIPELINE_ID)&statusFilter=completed&resultFilter=succeeded&\$top=1&api-version=7.1"
         CK_BUILD_ID=$(curl -s $LATEST_BUILD_URL | jq '.value[0].id')
         echo "Found latest CK build ID: $CK_BUILD_ID"
-        AZURE_URL="$AZ_API/build/builds/$CK_BUILD_ID/artifacts?artifactName=$ARTIFACT_NAME&api-version=7.1"
+        AZURE_URL="$AZ_API/build/builds/$CK_BUILD_ID/artifacts?api-version=7.1"
-        ARTIFACT_URL=$(curl -s $AZURE_URL | jq '.resource.downloadUrl' | tr -d '"')
+        ARTIFACT_URL=$(curl -s $AZURE_URL | \
+          jq --arg os "ubuntu2204" --arg gfx "${{ parameters.gpuTarget }}" '
+            .value
+            | map(select(.name | test($os) and test($gfx)))
+            | max_by(.name | capture("_(?<dropNumber>\\d+)").dropNumber | tonumber)
+            | .resource.downloadUrl
+          ' | \
+          tr -d '"')
         EXIT_CODE=2
       fi
 
@@ -57,8 +70,8 @@ steps:
       wget --tries=5 --waitretry=10 --retry-connrefused -nv $ARTIFACT_URL -O $(System.ArtifactsDirectory)/ck.zip
       unzip $(System.ArtifactsDirectory)/ck.zip -d $(System.ArtifactsDirectory)
       mkdir -p $(Agent.BuildDirectory)/rocm
-      tar -zxvf $(System.ArtifactsDirectory)/$ARTIFACT_NAME/*.tar.gz -C $(Agent.BuildDirectory)/rocm
+      tar -zxvf $(System.ArtifactsDirectory)/composable_kernel*/*.tar.gz -C $(Agent.BuildDirectory)/rocm
-      rm -r $(System.ArtifactsDirectory)/ck.zip $(System.ArtifactsDirectory)/$ARTIFACT_NAME
+      rm -r $(System.ArtifactsDirectory)/ck.zip $(System.ArtifactsDirectory)/composable_kernel*
 
       if [[ $EXIT_CODE -ne 0 ]]; then
         BUILD_COMMIT=$(curl -s $AZ_API/build/builds/$CK_BUILD_ID | jq '.sourceVersion' | tr -d '"')

.azuredevops/variables-global.yml

@@ -66,11 +66,11 @@ variables:
 - name: HIP_TESTS_PIPELINE_ID
   value: 233
 - name: HIPBLAS_COMMON_PIPELINE_ID
-  value: 223
+  value: 300
 - name: HIPBLAS_PIPELINE_ID
   value: 87
 - name: HIPBLASLT_PIPELINE_ID
-  value: 112
+  value: 301
 - name: HIPCUB_PIPELINE_ID
   value: 277
 - name: HIPFFT_PIPELINE_ID
@@ -86,7 +86,7 @@ variables:
 - name: HIPSPARSE_PIPELINE_ID
   value: 83
 - name: HIPSPARSELT_PIPELINE_ID
-  value: 104
+  value: 309
 - name: HIPTENSOR_PIPELINE_ID
   value: 105
 - name: LLVM_PROJECT_PIPELINE_ID
@@ -104,7 +104,7 @@ variables:
 - name: ROCALUTION_PIPELINE_ID
   value: 89
 - name: ROCBLAS_PIPELINE_ID
-  value: 85
+  value: 302
 - name: ROCDBGAPI_PIPELINE_ID
   value: 135
 - name: ROCDECODE_PIPELINE_ID
@@ -154,7 +154,7 @@ variables:
 - name: ROCSOLVER_PIPELINE_ID
   value: 81
 - name: ROCSPARSE_PIPELINE_ID
-  value: 98
+  value: 314
 - name: ROCTHRUST_PIPELINE_ID
   value: 276
 - name: ROCTRACER_PIPELINE_ID

.wordlist.txt

@@ -721,11 +721,13 @@ linearized
 linter
 linux
 llvm
+lm
 localscratch
 logits
 lossy
 macOS
 matchers
+megatron
 microarchitecture
 migraphx
 migratable
@@ -797,6 +799,7 @@ quantile
 quantizer
 quasirandom
 queueing
+qwen
 radeon
 rccl
 rdc

CHANGELOG.md

@@ -190,9 +190,8 @@ for a complete overview of this release.
 
 * When using the `--follow` flag with `amd-smi ras --cper`, CPER entries are not streamed continuously as intended. This will be fixed in an upcoming ROCm release.
 
-```{note}
+> [!NOTE]
-See the full [AMD SMI changelog](https://github.com/ROCm/amdsmi/blob/release/rocm-rel-6.4/CHANGELOG.md) for details, examples, and in-depth descriptions.
+> See the full [AMD SMI changelog](https://github.com/ROCm/amdsmi/blob/release/rocm-rel-6.4/CHANGELOG.md) for details, examples, and in-depth descriptions.
-```
 
 ### **HIP** (6.4.1)
 
@@ -273,9 +272,8 @@ See the full [AMD SMI changelog](https://github.com/ROCm/amdsmi/blob/release/roc
 
 - Fixed partition enumeration. It now refers to the correct DRM Render and Card paths.
 
-```{note}
+> [!NOTE]
-See the full [ROCm SMI changelog](https://github.com/ROCm/rocm_smi_lib/blob/release/rocm-rel-6.4/CHANGELOG.md) for details, examples, and in-depth descriptions.
+> See the full [ROCm SMI changelog](https://github.com/ROCm/rocm_smi_lib/blob/release/rocm-rel-6.4/CHANGELOG.md) for details, examples, and in-depth descriptions.
-```
 
 ### **ROCm Systems Profiler** (1.0.1)
 
@@ -413,9 +411,8 @@ Some workaround options are as follows:
 
 - The `pasid` field in struct `amdsmi_process_info_t` will be deprecated in a future ROCm release.
 
-```{note}
+> [!NOTE]
-See the full [AMD SMI changelog](https://github.com/ROCm/amdsmi/blob/release/rocm-rel-6.4/CHANGELOG.md) for details, examples, and in-depth descriptions.
+> See the full [AMD SMI changelog](https://github.com/ROCm/amdsmi/blob/release/rocm-rel-6.4/CHANGELOG.md) for details, examples, and in-depth descriptions.
-```
 
 ### **AMDMIGraphX** (2.12.0)
 
@@ -1023,9 +1020,8 @@ The following lists the backward incompatible changes planned for upcoming major
 
 - Fixed `rsmi_dev_target_graphics_version_get`, `rocm-smi --showhw`, and `rocm-smi --showprod` not displaying graphics version correctly for Instinct MI200 series, MI100 series, and RDNA3-based GPUs. 
 
-```{note}
+> [!NOTE]
-See the full [ROCm SMI changelog](https://github.com/ROCm/rocm_smi_lib/blob/release/rocm-rel-6.4/CHANGELOG.md) for details, examples, and in-depth descriptions.
+> See the full [ROCm SMI changelog](https://github.com/ROCm/rocm_smi_lib/blob/release/rocm-rel-6.4/CHANGELOG.md) for details, examples, and in-depth descriptions.
-```
 
 ### **ROCm Systems Profiler** (1.0.0)
 
@@ -1451,9 +1447,8 @@ for a complete overview of this release.
 * Fixed `amd-smi monitor`'s reporting of encode and decode information. `VCLOCK` and `DCLOCK` are
   now associated with both `ENC_UTIL` and `DEC_UTIL`.
 
-```{note}
+> [!NOTE]
-See the full [AMD SMI changelog](https://github.com/ROCm/amdsmi/blob/6.3.x/CHANGELOG.md) for more details and examples.
+> See the full [AMD SMI changelog](https://github.com/ROCm/amdsmi/blob/6.3.x/CHANGELOG.md) for more details and examples.
-```
 
 ### **HIP** (6.3.1)
 
@@ -1657,9 +1652,8 @@ for a complete overview of this release.
   - The new partition command can display GPU information, including memory and accelerator partition information.
   - The command will be at full functionality once additional partition information from `amdsmi_get_gpu_accelerator_partition_profile()` has been implemented.
 
-```{note}
+> [!NOTE]
-See the full [AMD SMI changelog](https://github.com/ROCm/amdsmi/blob/6.3.x/CHANGELOG.md) for more details and examples.
+> See the full [AMD SMI changelog](https://github.com/ROCm/amdsmi/blob/6.3.x/CHANGELOG.md) for more details and examples.
-```
 
 ### **HIP** (6.3.0)
 
@@ -1793,18 +1787,17 @@ See the full [AMD SMI changelog](https://github.com/ROCm/amdsmi/blob/6.3.x/CHANG
 
 * Support for `fp8` data types
 
-### **hipRAND** (2.11.0[*](#id22))
+### **hipRAND** (2.11.0)
+
+> [!NOTE]
+> In ROCm 6.3.0, the hipRAND package version is incorrectly set to `2.11.0`.
+> In ROCm 6.2.4, the hipRAND package version was `2.11.1`.
+> The hipRAND version number will be corrected in a future ROCm release.
 
 #### Changed
 
 * Updated the default value for the `-a` argument from `rmake.py` to `gfx906:xnack-,gfx1030,gfx1100,gfx1101,gfx1102`.
 
-#### Known issues
-
-* In ROCm 6.3.0, the hipRAND package version is incorrectly set to `2.11.0`. In ROCm
-  6.2.4, the hipRAND package version was `2.11.1`. The hipRAND version number will be corrected in a
-  future ROCm release.
-
 #### Resolved issues
 
 * Fixed an issue in `rmake.py` where the list storing the CMake options would contain individual characters instead of a full string of options.
@@ -2005,7 +1998,7 @@ See the full [AMD SMI changelog](https://github.com/ROCm/amdsmi/blob/6.3.x/CHANG
 
 #### Known issues
 
-* See [MIVisionX memory access fault in Canny edge detection](#mivisionx-memory-access-fault-in-canny-edge-detection).
+* See [MIVisionX memory access fault in Canny edge detection](https://github.com/ROCm/ROCm/issues/4086).
 * Package installation requires the manual installation of OpenCV.
 * Installation on CentOS/RedHat/SLES requires the manual installation of the `FFMPEG Dev` package.
 * Hardware decode requires installation with `--usecase=graphics` in addition to `--usecase=rocm`.
@@ -2196,9 +2189,9 @@ See the full [AMD SMI changelog](https://github.com/ROCm/amdsmi/blob/6.3.x/CHANG
 
 #### Known issues
 
-- See [ROCm Compute Profiler post-upgrade](#rocm-compute-profiler-post-upgrade).
+- See [ROCm Compute Profiler post-upgrade](https://github.com/ROCm/ROCm/issues/4082).
 
-- See [ROCm Compute Profiler CTest failure in CI](#rocm-compute-profiler-ctest-failure-in-ci).
+- See [ROCm Compute Profiler CTest failure in CI](https://github.com/ROCm/ROCm/issues/4085).
 
 ### **ROCm Data Center Tool** (0.3.0)
 
@@ -2211,7 +2204,7 @@ See the full [AMD SMI changelog](https://github.com/ROCm/amdsmi/blob/6.3.x/CHANG
 
 #### Known issues
 
-- See [ROCm Data Center Tool incorrect RHEL9 package version](#rocm-data-center-tool-incorrect-rhel9-package-version).
+- See [ROCm Data Center Tool incorrect RHEL9 package version](https://github.com/ROCm/ROCm/issues/4089).
 
 ### **ROCm SMI** (7.4.0)
 
@@ -2249,9 +2242,8 @@ memory partition modes upon an invalid argument return from memory partition mod
 
 - C++ tests for `memorypartition_read_write` are to be re-enabled in a future ROCm release.
 
-```{note}
+> [!NOTE]
-See the full [ROCm SMI changelog](https://github.com/ROCm/rocm_smi_lib/blob/6.3.x/CHANGELOG.md) for more details and examples.
+> See the full [ROCm SMI changelog](https://github.com/ROCm/rocm_smi_lib/blob/6.3.x/CHANGELOG.md) for more details and examples.
-```
 
 ### **ROCm Systems Profiler** (0.1.0)
 
@@ -2265,7 +2257,7 @@ See the full [ROCm SMI changelog](https://github.com/ROCm/rocm_smi_lib/blob/6.3.
 
 #### Known issues
 
-- See [ROCm Systems Profiler post-upgrade](#rocm-systems-profiler-post-upgrade).
+- See [ROCm Systems Profiler post-upgrade](https://github.com/ROCm/ROCm/issues/4083).
 
 ### **ROCm Validation Suite** (1.1.0)
 
@@ -2279,7 +2271,7 @@ See the full [ROCm SMI changelog](https://github.com/ROCm/rocm_smi_lib/blob/6.3.
 
 #### Known issues
 
-- See [ROCm Validation Suite needs specified configuration file](#rocm-validation-suite-needs-specified-configuration-file).
+- See [ROCm Validation Suite needs specified configuration file](https://github.com/ROCm/ROCm/issues/4090).
 
 ### **rocPRIM** (3.3.0)
 
@@ -3022,10 +3014,8 @@ for a complete overview of this release.
 
 See [issue #3500](https://github.com/ROCm/ROCm/issues/3500) on GitHub.
 
-```{note}
+> [!NOTE]
-See the [detailed AMD SMI changelog](https://github.com/ROCm/amdsmi/blob/docs/6.2.0/CHANGELOG.md)
+> See the [detailed AMD SMI changelog](https://github.com/ROCm/amdsmi/blob/docs/6.2.0/CHANGELOG.md) on GitHub for more information.
-on GitHub for more information.
-```
 
 ### **Composable Kernel** (1.1.0)
 
@@ -3624,9 +3614,8 @@ The compiler may incorrectly compile a program that uses the
 the function is undefined along some path to the function. For most functions,
 uninitialized inputs cause undefined behavior.
 
-```{note}
+> [!NOTE]
-The ``-Wall`` compilation flag prompts the compiler to generate a warning if a variable is uninitialized along some path.
+> The ``-Wall`` compilation flag prompts the compiler to generate a warning if a variable is uninitialized along some path.
-```
 
 As a workaround, initialize the parameters to ``__shfl``. For example:
 
@@ -3947,10 +3936,8 @@ See [issue #3498](https://github.com/ROCm/ROCm/issues/3498) on GitHub.
 
 - Fixed Partition ID CLI output.
 
-```{note}
+> [!NOTE]
-See the [detailed ROCm SMI changelog](https://github.com/ROCm/rocm_smi_lib/blob/docs/6.2.0/CHANGELOG.md)
+> See the [detailed ROCm SMI changelog](https://github.com/ROCm/rocm_smi_lib/blob/docs/6.2.0/CHANGELOG.md) on GitHub for more information.
-on GitHub for more information.
-```
 
 ### **ROCm Validation Suite** (1.0.0)
 
@@ -4320,9 +4307,8 @@ for a complete overview of this release.
 * Fixed the `amdsmitstReadWrite.TestPowerCapReadWrite` test for RDNA3, RDNA2, and MI100 devices.
 * Fixed an issue with the `amdsmi_get_gpu_memory_reserved_pages` and `amdsmi_get_gpu_bad_page_info` Python interface calls.
 
-```{note}
+> [!NOTE]
-See the AMD SMI [detailed changelog](https://github.com/ROCm/amdsmi/blob/rocm-6.1.x/CHANGELOG.md) with code samples for more information.
+> See the AMD SMI [detailed changelog](https://github.com/ROCm/amdsmi/blob/rocm-6.1.x/CHANGELOG.md) with code samples for more information.
-```
 
 ### **RCCL** (2.18.6)
 
@@ -4402,9 +4388,8 @@ for a complete overview of this release.
 
 - `amd-smi bad-pages` can result in a `ValueError: Null pointer access` error when using some PMU firmware versions.
 
-```{note}
+> [!NOTE]
-See the [detailed changelog](https://github.com/ROCm/amdsmi/blob/docs/6.1.1/CHANGELOG.md) with code samples for more information.
+> See the [detailed changelog](https://github.com/ROCm/amdsmi/blob/docs/6.1.1/CHANGELOG.md) with code samples for more information.
-```
 
 ### **hipBLASLt** (0.7.0)
 
@@ -4473,9 +4458,8 @@ See the [detailed changelog](https://github.com/ROCm/amdsmi/blob/docs/6.1.1/CHAN
 
 - ROCm SMI reports GPU utilization incorrectly for RDNA3 GPUs in some situations. See the issue on [GitHub](https://github.com/ROCm/ROCm/issues/3112).
 
-```{note}
+> [!NOTE]
-See the [detailed ROCm SMI changelog](https://github.com/ROCm/rocm_smi_lib/blob/docs/6.1.1/CHANGELOG.md) with code samples for more information.
+> See the [detailed ROCm SMI changelog](https://github.com/ROCm/rocm_smi_lib/blob/docs/6.1.1/CHANGELOG.md) with code samples for more information.
-```
 
 ## ROCm 6.1.0
 
@@ -5192,16 +5176,16 @@ on GitHub for a complete overview of this release.
 
 ### **rocSPARSE** (2.5.4)
 
-##### Added
+#### Added
 
 - Added more mixed precisions for SpMV, (matrix: float, vectors: double, calculation: double) and (matrix: rocsparse_float_complex, vectors: rocsparse_double_complex, calculation: rocsparse_double_complex)
 - Added support for gfx940, gfx941 and gfx942
 
-##### Optimized
+#### Optimized
 
 - Fixed a bug in csrsm and bsrsm
 
-##### Known issues
+#### Known issues
 
 In csritlu0, the algorithm rocsparse_itilu0_alg_sync_split_fusion has some accuracy issues to investigate with XNACK enabled. The fallback is rocsparse_itilu0_alg_sync_split.
 
@@ -5287,7 +5271,7 @@ on GitHub for a complete overview of this release.
 
 ### **HIP** (5.6.0)
 
-##### Added
+#### Added
 
 - Added hipRTC support for amd_hip_fp16
 - Added hipStreamGetDevice implementation to get the device associated with the stream
@@ -5296,7 +5280,7 @@ on GitHub for a complete overview of this release.
 - hipArrayGetDescriptor for getting 1D or 2D array descriptor
 - hipArray3DGetDescriptor to get 3D array descriptor
 
-##### Changed
+#### Changed
 
 - hipMallocAsync to return success for zero size allocation to match hipMalloc
 - Separation of hipcc perl binaries from HIP project to hipcc project. hip-devel package depends on newly added hipcc package
@@ -5601,15 +5585,15 @@ $ gcc main.c -I/opt/rocm-5.6.0/include -L/opt/rocm-5.6.0/lib -lrocprofiler64-v2
 The resulting `a.out` will depend on
 `/opt/rocm-5.6.0/lib/librocprofiler64.so.2`.
 
-##### Added
+#### Added
 
 - 'end_time' need to be disabled in roctx_trace.txt
 
-##### Optimized
+#### Optimized
 
 - Improved Test Suite
 
-##### Resolved issues
+#### Resolved issues
 
 - rocprof in ROcm/5.4.0 gpu selector broken.
 - rocprof in ROCm/5.4.1 fails to generate kernel info.

docs/compatibility/compatibility-matrix-historical-6.0.csv

@@ -30,6 +30,9 @@ ROCm Version,6.4.2,6.4.1,6.4.0,6.3.3,6.3.2,6.3.1,6.3.0,6.2.4,6.2.2,6.2.1,6.2.0,
       :doc:`PyTorch <../compatibility/ml-compatibility/pytorch-compatibility>`,"2.6, 2.5, 2.4, 2.3","2.6, 2.5, 2.4, 2.3","2.6, 2.5, 2.4, 2.3","2.4, 2.3, 2.2, 1.13","2.4, 2.3, 2.2, 1.13","2.4, 2.3, 2.2, 1.13","2.4, 2.3, 2.2, 2.1, 2.0, 1.13","2.3, 2.2, 2.1, 2.0, 1.13","2.3, 2.2, 2.1, 2.0, 1.13","2.3, 2.2, 2.1, 2.0, 1.13","2.3, 2.2, 2.1, 2.0, 1.13","2.1, 2.0, 1.13","2.1, 2.0, 1.13","2.1, 2.0, 1.13","2.1, 2.0, 1.13","2.1, 2.0, 1.13","2.1, 2.0, 1.13"
       :doc:`TensorFlow <../compatibility/ml-compatibility/tensorflow-compatibility>`,"2.18.1, 2.17.1, 2.16.2","2.18.1, 2.17.1, 2.16.2","2.18.1, 2.17.1, 2.16.2","2.17.0, 2.16.2, 2.15.1","2.17.0, 2.16.2, 2.15.1","2.17.0, 2.16.2, 2.15.1","2.17.0, 2.16.2, 2.15.1","2.16.1, 2.15.1, 2.14.1","2.16.1, 2.15.1, 2.14.1","2.16.1, 2.15.1, 2.14.1","2.16.1, 2.15.1, 2.14.1","2.15.0, 2.14.0, 2.13.1","2.15.0, 2.14.0, 2.13.1","2.15.0, 2.14.0, 2.13.1","2.15.0, 2.14.0, 2.13.1","2.14.0, 2.13.1, 2.12.1","2.14.0, 2.13.1, 2.12.1"
       :doc:`JAX <../compatibility/ml-compatibility/jax-compatibility>`,0.4.35,0.4.35,0.4.35,0.4.31,0.4.31,0.4.31,0.4.31,0.4.26,0.4.26,0.4.26,0.4.26,0.4.26,0.4.26,0.4.26,0.4.26,0.4.26,0.4.26
+      :doc:`Stanford Megatron-LM <../compatibility/ml-compatibility/stanford-megatron-lm-compatibility>`,N/A,N/A,N/A,N/A,N/A,85f95ae,N/A,N/A,N/A,N/A,N/A,N/A,N/A,N/A,N/A,N/A
+      :doc:`DGL <../compatibility/ml-compatibility/dgl-compatibility>`,2.4.0,2.4.0,N/A,N/A,N/A,N/A,N/A,N/A,N/A,N/A,N/A,N/A,N/A,N/A,N/A,N/A
+      :doc:`verl <../compatibility/ml-compatibility/verl-compatibility>` [#verl_compat]_,N/A,N/A,N/A,N/A,N/A,N/A,N/A,N/A,N/A,0.3.0.post0,N/A,N/A,N/A,N/A,N/A,N/A
       `ONNX Runtime <https://onnxruntime.ai/docs/build/eps.html#amd-migraphx>`_,1.2,1.2,1.2,1.17.3,1.17.3,1.17.3,1.17.3,1.17.3,1.17.3,1.17.3,1.17.3,1.17.3,1.17.3,1.17.3,1.17.3,1.14.1,1.14.1
 ,,,,,,,,,,,,,,,,,
       ,,,,,,,,,,,,,,,,,

docs/compatibility/compatibility-matrix.rst

@@ -54,8 +54,8 @@ compatibility and system requirements.
       FRAMEWORK SUPPORT,.. _framework-support-compatibility-matrix:,,
       :doc:`PyTorch <../compatibility/ml-compatibility/pytorch-compatibility>`,"2.6, 2.5, 2.4, 2.3","2.6, 2.5, 2.4, 2.3","2.4, 2.3, 2.2, 2.1, 2.0, 1.13"
       :doc:`TensorFlow <../compatibility/ml-compatibility/tensorflow-compatibility>`,"2.18.1, 2.17.1, 2.16.2","2.18.1, 2.17.1, 2.16.2","2.17.0, 2.16.2, 2.15.1"
-      :doc:`JAX <../compatibility/ml-compatibility/jax-compatibility>`,0.4.35,0.4.35,0.4.31
+      :doc:`JAX <../compatibility/ml-compatibility/jax-compatibility>`,0.4.35,0.4.35,0.4.31  
-      :doc:`Stanford Megatron-LM <../compatibility/ml-compatibility/stanford-megatron-lm-compatibility>`,N/A,N/A,`85f95ae <https://github.com/stanford-futuredata/Megatron-LM/commit/85f95aef3b648075fe6f291c86714fdcbd9cd1f5>`_
+      :doc:`Stanford Megatron-LM <../compatibility/ml-compatibility/stanford-megatron-lm-compatibility>`,N/A,N/A,85f95ae
       :doc:`DGL <../compatibility/ml-compatibility/dgl-compatibility>`,2.4.0,2.4.0,N/A
       `ONNX Runtime <https://onnxruntime.ai/docs/build/eps.html#amd-migraphx>`_,1.2,1.2,1.17.3
       ,,,

docs/compatibility/ml-compatibility/pytorch-compatibility.rst

@@ -10,10 +10,10 @@
 PyTorch compatibility
 ********************************************************************************
 
-`PyTorch <https://pytorch.org/>`_ is an open-source tensor library designed for
+`PyTorch <https://pytorch.org/>`__ is an open-source tensor library designed for
 deep learning. PyTorch on ROCm provides mixed-precision and large-scale training
-using `MIOpen <https://github.com/ROCm/MIOpen>`_ and
+using `MIOpen <https://github.com/ROCm/MIOpen>`__ and
-`RCCL <https://github.com/ROCm/rccl>`_ libraries.
+`RCCL <https://github.com/ROCm/rccl>`__ libraries.
 
 ROCm support for PyTorch is upstreamed into the official PyTorch repository. Due
 to independent compatibility considerations, this results in two distinct
@@ -27,7 +27,7 @@ release cycles for PyTorch on ROCm:
   - Offers :ref:`Docker images <pytorch-docker-compat>` with ROCm and PyTorch
     preinstalled.
 
-  - ROCm PyTorch repository: `<https://github.com/ROCm/pytorch>`_
+  - ROCm PyTorch repository: `<https://github.com/ROCm/pytorch>`__
 
   - See the :doc:`ROCm PyTorch installation guide <rocm-install-on-linux:install/3rd-party/pytorch-install>`
     to get started.
@@ -37,10 +37,10 @@ release cycles for PyTorch on ROCm:
   - Provides the latest stable version of PyTorch  but might not necessarily
     support the latest ROCm version.
 
-  - Official PyTorch repository: `<https://github.com/pytorch/pytorch>`_
+  - Official PyTorch repository: `<https://github.com/pytorch/pytorch>`__
 
-  - See the `Nightly and latest stable version installation guide <https://pytorch.org/get-started/locally/>`_
+  - See the `Nightly and latest stable version installation guide <https://pytorch.org/get-started/locally/>`__
-    or `Previous versions <https://pytorch.org/get-started/previous-versions/>`_
+    or `Previous versions <https://pytorch.org/get-started/previous-versions/>`__
     to get started.
 
 PyTorch includes tooling that generates HIP source code from the CUDA backend.
@@ -82,7 +82,7 @@ Use cases and recommendations
   use of PyTorch on the ROCm platform and focuses on efficiently leveraging AMD
   GPU hardware for training and inference tasks in AI applications.
 
-For more use cases and recommendations, see `ROCm PyTorch blog posts <https://rocm.blogs.amd.com/blog/tag/pytorch.html>`_.
+For more use cases and recommendations, see `ROCm PyTorch blog posts <https://rocm.blogs.amd.com/blog/tag/pytorch.html>`__.
 
 .. _pytorch-docker-compat:
 
@@ -93,9 +93,9 @@ Docker image compatibility
 
    <i class="fab fa-docker"></i>
 
-AMD validates and publishes `PyTorch images <https://hub.docker.com/r/rocm/pytorch>`_
+AMD validates and publishes `PyTorch images <https://hub.docker.com/r/rocm/pytorch>`__
 with ROCm backends on Docker Hub. The following Docker image tags and associated
-inventories were tested on `ROCm 6.4.1 <https://repo.radeon.com/rocm/apt/6.4.1/>`_.
+inventories were tested on `ROCm 6.4.1 <https://repo.radeon.com/rocm/apt/6.4.1/>`__.
 Click |docker-icon| to view the image on Docker Hub.
 
 .. list-table:: PyTorch Docker image components
@@ -118,121 +118,121 @@ Click |docker-icon| to view the image on Docker Hub.
 
            <a href="https://hub.docker.com/layers/rocm/pytorch/rocm6.4.1_ubuntu24.04_py3.12_pytorch_release_2.6.0/images/sha256-c76af9bfb1c25b0f40d4c29e8652105c57250bf018d23ff595b06bd79666fdd7"><i class="fab fa-docker fa-lg"></i></a>
 
-      - `2.6.0 <https://github.com/ROCm/pytorch/tree/release/2.6>`_
+      - `2.6.0 <https://github.com/ROCm/pytorch/tree/release/2.6>`__
       - 24.04
-      - `3.12.10 <https://www.python.org/downloads/release/python-31210/>`_
+      - `3.12.10 <https://www.python.org/downloads/release/python-31210/>`__
-      - `1.6.0 <https://github.com/ROCm/apex/tree/release/1.6.0>`_
+      - `1.6.0 <https://github.com/ROCm/apex/tree/release/1.6.0>`__
-      - `0.21.0 <https://github.com/pytorch/vision/tree/v0.21.0>`_
+      - `0.21.0 <https://github.com/pytorch/vision/tree/v0.21.0>`__
-      - `2.13.0 <https://github.com/tensorflow/tensorboard/tree/2.13.0>`_
+      - `2.13.0 <https://github.com/tensorflow/tensorboard/tree/2.13.0>`__
-      - `master <https://bitbucket.org/icl/magma/src/master/>`_
+      - `master <https://bitbucket.org/icl/magma/src/master/>`__
-      - `1.16.0 <https://github.com/openucx/ucx/tree/v1.16.0>`_
+      - `1.16.0 <https://github.com/openucx/ucx/tree/v1.16.0>`__
-      - `4.1.6-7ubuntu2 <https://github.com/open-mpi/ompi/tree/v4.1.6>`_
+      - `4.1.6-7ubuntu2 <https://github.com/open-mpi/ompi/tree/v4.1.6>`__
-      - `5.3-1.0.5.0 <https://content.mellanox.com/ofed/MLNX_OFED-5.3-1.0.5.0/MLNX_OFED_LINUX-5.3-1.0.5.0-ubuntu20.04-x86_64.tgz>`_
+      - `5.3-1.0.5.0 <https://content.mellanox.com/ofed/MLNX_OFED-5.3-1.0.5.0/MLNX_OFED_LINUX-5.3-1.0.5.0-ubuntu20.04-x86_64.tgz>`__
 
     * - .. raw:: html
 
            <a href="https://hub.docker.com/layers/rocm/pytorch/rocm6.4.1_ubuntu22.04_py3.10_pytorch_release_2.6.0/images/sha256-f9d226135d51831c810dcb1251636ec61f85c65fcdda03e188c053a5d4f6585b"><i class="fab fa-docker fa-lg"></i></a>
 
-      - `2.6.0 <https://github.com/ROCm/pytorch/tree/release/2.6>`_
+      - `2.6.0 <https://github.com/ROCm/pytorch/tree/release/2.6>`__
       - 22.04
-      - `3.10.17 <https://www.python.org/downloads/release/python-31017/>`_
+      - `3.10.17 <https://www.python.org/downloads/release/python-31017/>`__
-      - `1.6.0 <https://github.com/ROCm/apex/tree/release/1.6.0>`_
+      - `1.6.0 <https://github.com/ROCm/apex/tree/release/1.6.0>`__
-      - `0.21.0 <https://github.com/pytorch/vision/tree/v0.21.0>`_
+      - `0.21.0 <https://github.com/pytorch/vision/tree/v0.21.0>`__
-      - `2.13.0 <https://github.com/tensorflow/tensorboard/tree/2.13.0>`_
+      - `2.13.0 <https://github.com/tensorflow/tensorboard/tree/2.13.0>`__
-      - `master <https://bitbucket.org/icl/magma/src/master/>`_
+      - `master <https://bitbucket.org/icl/magma/src/master/>`__
-      - `1.12.1~rc2-1 <https://github.com/openucx/ucx/tree/v1.12.1>`_
+      - `1.12.1~rc2-1 <https://github.com/openucx/ucx/tree/v1.12.1>`__
-      - `4.1.2-2ubuntu1 <https://github.com/open-mpi/ompi/tree/v4.1.2>`_
+      - `4.1.2-2ubuntu1 <https://github.com/open-mpi/ompi/tree/v4.1.2>`__
-      - `5.3-1.0.5.0 <https://content.mellanox.com/ofed/MLNX_OFED-5.3-1.0.5.0/MLNX_OFED_LINUX-5.3-1.0.5.0-ubuntu20.04-x86_64.tgz>`_
+      - `5.3-1.0.5.0 <https://content.mellanox.com/ofed/MLNX_OFED-5.3-1.0.5.0/MLNX_OFED_LINUX-5.3-1.0.5.0-ubuntu20.04-x86_64.tgz>`__
 
     * - .. raw:: html
 
            <a href="https://hub.docker.com/layers/rocm/pytorch/rocm6.4.1_ubuntu24.04_py3.12_pytorch_release_2.5.1/images/sha256-3490e74d4f43dcdb3351dd334108d1ccd47e5a687c0523a2424ac1bcdd3dd6dd"><i class="fab fa-docker fa-lg"></i></a>
 
-      - `2.5.1 <https://github.com/ROCm/pytorch/tree/release/2.5>`_
+      - `2.5.1 <https://github.com/ROCm/pytorch/tree/release/2.5>`__
       - 24.04
-      - `3.12.10 <https://www.python.org/downloads/release/python-31210/>`_
+      - `3.12.10 <https://www.python.org/downloads/release/python-31210/>`__
-      - `1.5.0 <https://github.com/ROCm/apex/tree/release/1.5.0>`_
+      - `1.5.0 <https://github.com/ROCm/apex/tree/release/1.5.0>`__
-      - `0.20.1 <https://github.com/pytorch/vision/tree/v0.20.1>`_
+      - `0.20.1 <https://github.com/pytorch/vision/tree/v0.20.1>`__
-      - `2.13.0 <https://github.com/tensorflow/tensorboard/tree/2.13.0>`_
+      - `2.13.0 <https://github.com/tensorflow/tensorboard/tree/2.13.0>`__
-      - `master <https://bitbucket.org/icl/magma/src/master/>`_
+      - `master <https://bitbucket.org/icl/magma/src/master/>`__
-      - `1.16.0+ds-5ubuntu1 <https://github.com/openucx/ucx/tree/v1.10.0>`_
+      - `1.16.0+ds-5ubuntu1 <https://github.com/openucx/ucx/tree/v1.10.0>`__
-      - `4.1.6-7ubuntu2 <https://github.com/open-mpi/ompi/tree/v4.1.6>`_
+      - `4.1.6-7ubuntu2 <https://github.com/open-mpi/ompi/tree/v4.1.6>`__
-      - `5.3-1.0.5.0 <https://content.mellanox.com/ofed/MLNX_OFED-5.3-1.0.5.0/MLNX_OFED_LINUX-5.3-1.0.5.0-ubuntu20.04-x86_64.tgz>`_
+      - `5.3-1.0.5.0 <https://content.mellanox.com/ofed/MLNX_OFED-5.3-1.0.5.0/MLNX_OFED_LINUX-5.3-1.0.5.0-ubuntu20.04-x86_64.tgz>`__
 
     * - .. raw:: html
 
            <a href="https://hub.docker.com/layers/rocm/pytorch/rocm6.4.1_ubuntu22.04_py3.10_pytorch_release_2.5.1/images/sha256-26c5dfffb4a54625884abca83166940f17dd27bc75f1b24f6e80fbcb7d4e9afb"><i class="fab fa-docker fa-lg"></i></a>
 
-      - `2.5.1 <https://github.com/ROCm/pytorch/tree/release/2.5>`_
+      - `2.5.1 <https://github.com/ROCm/pytorch/tree/release/2.5>`__
       - 22.04
-      - `3.10.17 <https://www.python.org/downloads/release/python-31017/>`_
+      - `3.10.17 <https://www.python.org/downloads/release/python-31017/>`__
-      - `1.5.0 <https://github.com/ROCm/apex/tree/release/1.5.0>`_
+      - `1.5.0 <https://github.com/ROCm/apex/tree/release/1.5.0>`__
-      - `0.20.1 <https://github.com/pytorch/vision/tree/v0.20.1>`_
+      - `0.20.1 <https://github.com/pytorch/vision/tree/v0.20.1>`__
-      - `2.13.0 <https://github.com/tensorflow/tensorboard/tree/2.13.0>`_
+      - `2.13.0 <https://github.com/tensorflow/tensorboard/tree/2.13.0>`__
-      - `master <https://bitbucket.org/icl/magma/src/master/>`_
+      - `master <https://bitbucket.org/icl/magma/src/master/>`__
-      - `1.12.1~rc2-1 <https://github.com/openucx/ucx/tree/v1.12.1>`_
+      - `1.12.1~rc2-1 <https://github.com/openucx/ucx/tree/v1.12.1>`__
-      - `4.1.2-2ubuntu1 <https://github.com/open-mpi/ompi/tree/v4.1.2>`_
+      - `4.1.2-2ubuntu1 <https://github.com/open-mpi/ompi/tree/v4.1.2>`__
-      - `5.3-1.0.5.0 <https://content.mellanox.com/ofed/MLNX_OFED-5.3-1.0.5.0/MLNX_OFED_LINUX-5.3-1.0.5.0-ubuntu20.04-x86_64.tgz>`_
+      - `5.3-1.0.5.0 <https://content.mellanox.com/ofed/MLNX_OFED-5.3-1.0.5.0/MLNX_OFED_LINUX-5.3-1.0.5.0-ubuntu20.04-x86_64.tgz>`__
 
     * - .. raw:: html
 
            <a href="https://hub.docker.com/layers/rocm/pytorch/rocm6.4.1_ubuntu24.04_py3.12_pytorch_release_2.4.1/images/sha256-f378a24561fa6efc178b6dc93fc7d82e5b93653ecd59c89d4476674d29e1284d"><i class="fab fa-docker fa-lg"></i></a>
 
-      - `2.4.1 <https://github.com/ROCm/pytorch/tree/release/2.4>`_
+      - `2.4.1 <https://github.com/ROCm/pytorch/tree/release/2.4>`__
       - 24.04
-      - `3.12.10 <https://www.python.org/downloads/release/python-31210/>`_
+      - `3.12.10 <https://www.python.org/downloads/release/python-31210/>`__
-      - `1.4.0 <https://github.com/ROCm/apex/tree/release/1.4.0>`_
+      - `1.4.0 <https://github.com/ROCm/apex/tree/release/1.4.0>`__
-      - `0.19.0 <https://github.com/pytorch/vision/tree/v0.19.0>`_
+      - `0.19.0 <https://github.com/pytorch/vision/tree/v0.19.0>`__
-      - `2.13.0 <https://github.com/tensorflow/tensorboard/tree/2.13.0>`_
+      - `2.13.0 <https://github.com/tensorflow/tensorboard/tree/2.13.0>`__
-      - `master <https://bitbucket.org/icl/magma/src/master/>`_
+      - `master <https://bitbucket.org/icl/magma/src/master/>`__
-      - `1.16.0+ds-5ubuntu1 <https://github.com/openucx/ucx/tree/v1.16.0>`_
+      - `1.16.0+ds-5ubuntu1 <https://github.com/openucx/ucx/tree/v1.16.0>`__
-      - `4.1.6-7ubuntu2 <https://github.com/open-mpi/ompi/tree/v4.1.6>`_
+      - `4.1.6-7ubuntu2 <https://github.com/open-mpi/ompi/tree/v4.1.6>`__
-      - `5.3-1.0.5.0 <https://content.mellanox.com/ofed/MLNX_OFED-5.3-1.0.5.0/MLNX_OFED_LINUX-5.3-1.0.5.0-ubuntu20.04-x86_64.tgz>`_
+      - `5.3-1.0.5.0 <https://content.mellanox.com/ofed/MLNX_OFED-5.3-1.0.5.0/MLNX_OFED_LINUX-5.3-1.0.5.0-ubuntu20.04-x86_64.tgz>`__
 
     * - .. raw:: html
 
            <a href="https://hub.docker.com/layers/rocm/pytorch/rocm6.4.1_ubuntu22.04_py3.10_pytorch_release_2.4.1/images/sha256-2308dbd0e650b7bf8d548575cbb6e2bdc021f9386384ce570da16d58ee684d22"><i class="fab fa-docker fa-lg"></i></a>
 
-      - `2.4.1 <https://github.com/ROCm/pytorch/tree/release/2.4>`_
+      - `2.4.1 <https://github.com/ROCm/pytorch/tree/release/2.4>`__
       - 22.04
-      - `3.10.17 <https://www.python.org/downloads/release/python-31017/>`_
+      - `3.10.17 <https://www.python.org/downloads/release/python-31017/>`__
-      - `1.4.0 <https://github.com/ROCm/apex/tree/release/1.4.0>`_
+      - `1.4.0 <https://github.com/ROCm/apex/tree/release/1.4.0>`__
-      - `0.19.0 <https://github.com/pytorch/vision/tree/v0.19.0>`_
+      - `0.19.0 <https://github.com/pytorch/vision/tree/v0.19.0>`__
-      - `2.13.0 <https://github.com/tensorflow/tensorboard/tree/2.13.0>`_
+      - `2.13.0 <https://github.com/tensorflow/tensorboard/tree/2.13.0>`__
-      - `master <https://bitbucket.org/icl/magma/src/master/>`_
+      - `master <https://bitbucket.org/icl/magma/src/master/>`__
-      - `1.12.1~rc2-1 <https://github.com/openucx/ucx/tree/v1.12.1>`_
+      - `1.12.1~rc2-1 <https://github.com/openucx/ucx/tree/v1.12.1>`__
-      - `4.1.2-2ubuntu1 <https://github.com/open-mpi/ompi/tree/v4.1.2>`_
+      - `4.1.2-2ubuntu1 <https://github.com/open-mpi/ompi/tree/v4.1.2>`__
-      - `5.3-1.0.5.0 <https://content.mellanox.com/ofed/MLNX_OFED-5.3-1.0.5.0/MLNX_OFED_LINUX-5.3-1.0.5.0-ubuntu20.04-x86_64.tgz>`_
+      - `5.3-1.0.5.0 <https://content.mellanox.com/ofed/MLNX_OFED-5.3-1.0.5.0/MLNX_OFED_LINUX-5.3-1.0.5.0-ubuntu20.04-x86_64.tgz>`__
 
     * - .. raw:: html
 
            <a href="https://hub.docker.com/layers/rocm/pytorch/rocm6.4.1_ubuntu24.04_py3.12_pytorch_release_2.3.0/images/sha256-eefd2ab019728f91f94c5e6a9463cb0ea900b3011458d18fe5d88e50c0b57d86"><i class="fab fa-docker fa-lg"></i></a>
 
-      - `2.3.0 <https://github.com/ROCm/pytorch/tree/release/2.3>`_
+      - `2.3.0 <https://github.com/ROCm/pytorch/tree/release/2.3>`__
       - 24.04
-      - `3.12.10 <https://www.python.org/downloads/release/python-31210/>`_
+      - `3.12.10 <https://www.python.org/downloads/release/python-31210/>`__
-      - `1.3.0 <https://github.com/ROCm/apex/tree/release/1.3.0>`_
+      - `1.3.0 <https://github.com/ROCm/apex/tree/release/1.3.0>`__
-      - `0.18.0 <https://github.com/pytorch/vision/tree/v0.18.0>`_
+      - `0.18.0 <https://github.com/pytorch/vision/tree/v0.18.0>`__
-      - `2.13.0 <https://github.com/tensorflow/tensorboard/tree/2.13>`_
+      - `2.13.0 <https://github.com/tensorflow/tensorboard/tree/2.13>`__
-      - `master <https://bitbucket.org/icl/magma/src/master/>`_
+      - `master <https://bitbucket.org/icl/magma/src/master/>`__
-      - `1.16.0+ds-5ubuntu1 <https://github.com/openucx/ucx/tree/v1.16.0>`_
+      - `1.16.0+ds-5ubuntu1 <https://github.com/openucx/ucx/tree/v1.16.0>`__
-      - `4.1.6-7ubuntu2 <https://github.com/open-mpi/ompi/tree/v4.1.6>`_
+      - `4.1.6-7ubuntu2 <https://github.com/open-mpi/ompi/tree/v4.1.6>`__
-      - `5.3-1.0.5.0 <https://content.mellanox.com/ofed/MLNX_OFED-5.3-1.0.5.0/MLNX_OFED_LINUX-5.3-1.0.5.0-ubuntu20.04-x86_64.tgz>`_
+      - `5.3-1.0.5.0 <https://content.mellanox.com/ofed/MLNX_OFED-5.3-1.0.5.0/MLNX_OFED_LINUX-5.3-1.0.5.0-ubuntu20.04-x86_64.tgz>`__
 
     * - .. raw:: html
 
            <a href="https://hub.docker.com/layers/rocm/pytorch/rocm6.4.1_ubuntu22.04_py3.10_pytorch_release_2.3.0/images/sha256-473643226ab0e93a04720b256ed772619878abf9c42b9f84828cefed522696fd"><i class="fab fa-docker fa-lg"></i></a>
 
-      - `2.3.0 <https://github.com/ROCm/pytorch/tree/release/2.3>`_
+      - `2.3.0 <https://github.com/ROCm/pytorch/tree/release/2.3>`__
       - 22.04
-      - `3.10.17 <https://www.python.org/downloads/release/python-31017/>`_
+      - `3.10.17 <https://www.python.org/downloads/release/python-31017/>`__
-      - `1.3.0 <https://github.com/ROCm/apex/tree/release/1.3.0>`_
+      - `1.3.0 <https://github.com/ROCm/apex/tree/release/1.3.0>`__
-      - `0.18.0 <https://github.com/pytorch/vision/tree/v0.18.0>`_
+      - `0.18.0 <https://github.com/pytorch/vision/tree/v0.18.0>`__
-      - `2.13.0 <https://github.com/tensorflow/tensorboard/tree/2.13>`_
+      - `2.13.0 <https://github.com/tensorflow/tensorboard/tree/2.13>`__
-      - `master <https://bitbucket.org/icl/magma/src/master/>`_
+      - `master <https://bitbucket.org/icl/magma/src/master/>`__
-      - `1.12.1~rc2-1 <https://github.com/openucx/ucx/tree/v1.12.1>`_
+      - `1.12.1~rc2-1 <https://github.com/openucx/ucx/tree/v1.12.1>`__
-      - `4.1.2-2ubuntu1 <https://github.com/open-mpi/ompi/tree/v4.1.2>`_
+      - `4.1.2-2ubuntu1 <https://github.com/open-mpi/ompi/tree/v4.1.2>`__
-      - `5.3-1.0.5.0 <https://content.mellanox.com/ofed/MLNX_OFED-5.3-1.0.5.0/MLNX_OFED_LINUX-5.3-1.0.5.0-ubuntu20.04-x86_64.tgz>`_
+      - `5.3-1.0.5.0 <https://content.mellanox.com/ofed/MLNX_OFED-5.3-1.0.5.0/MLNX_OFED_LINUX-5.3-1.0.5.0-ubuntu20.04-x86_64.tgz>`__
 
 Key ROCm libraries for PyTorch
 ================================================================================
@@ -248,121 +248,121 @@ feature set available to developers.
       - Version
       - Purpose
       - Used in
-    * - `Composable Kernel <https://github.com/ROCm/composable_kernel>`_
+    * - `Composable Kernel <https://github.com/ROCm/composable_kernel>`__
       - :version-ref:`"Composable Kernel" rocm_version`
       - Enables faster execution of core operations like matrix multiplication
         (GEMM), convolutions and transformations.
       - Speeds up ``torch.permute``, ``torch.view``, ``torch.matmul``,
         ``torch.mm``, ``torch.bmm``, ``torch.nn.Conv2d``, ``torch.nn.Conv3d``
         and ``torch.nn.MultiheadAttention``.
-    * - `hipBLAS <https://github.com/ROCm/hipBLAS>`_
+    * - `hipBLAS <https://github.com/ROCm/hipBLAS>`__
       - :version-ref:`hipBLAS rocm_version`
       - Provides GPU-accelerated Basic Linear Algebra Subprograms (BLAS) for
         matrix and vector operations.
       - Supports operations such as matrix multiplication, matrix-vector
         products, and tensor contractions. Utilized in both dense and batched
         linear algebra operations.
-    * - `hipBLASLt <https://github.com/ROCm/hipBLASLt>`_
+    * - `hipBLASLt <https://github.com/ROCm/hipBLASLt>`__
       - :version-ref:`hipBLASLt rocm_version`
       - hipBLASLt is an extension of the hipBLAS library, providing additional
         features like epilogues fused into the matrix multiplication kernel or
         use of integer tensor cores.
       - Accelerates operations such as ``torch.matmul``, ``torch.mm``, and the
         matrix multiplications used in convolutional and linear layers.
-    * - `hipCUB <https://github.com/ROCm/hipCUB>`_
+    * - `hipCUB <https://github.com/ROCm/hipCUB>`__
       - :version-ref:`hipCUB rocm_version`
       - Provides a C++ template library for parallel algorithms for reduction,
         scan, sort and select.
       - Supports operations such as ``torch.sum``, ``torch.cumsum``,
         ``torch.sort`` irregular shapes often involve scanning, sorting, and
         filtering, which hipCUB handles efficiently.
-    * - `hipFFT <https://github.com/ROCm/hipFFT>`_
+    * - `hipFFT <https://github.com/ROCm/hipFFT>`__
       - :version-ref:`hipFFT rocm_version`
       - Provides GPU-accelerated Fast Fourier Transform (FFT) operations.
       - Used in functions like the ``torch.fft`` module.
-    * - `hipRAND <https://github.com/ROCm/hipRAND>`_
+    * - `hipRAND <https://github.com/ROCm/hipRAND>`__
       - :version-ref:`hipRAND rocm_version`
       - Provides fast random number generation for GPUs.
       - The ``torch.rand``, ``torch.randn``, and stochastic layers like
         ``torch.nn.Dropout`` rely on hipRAND.
-    * - `hipSOLVER <https://github.com/ROCm/hipSOLVER>`_
+    * - `hipSOLVER <https://github.com/ROCm/hipSOLVER>`__
       - :version-ref:`hipSOLVER rocm_version`
       - Provides GPU-accelerated solvers for linear systems, eigenvalues, and
         singular value decompositions (SVD).
       - Supports functions like ``torch.linalg.solve``,
         ``torch.linalg.eig``, and ``torch.linalg.svd``.
-    * - `hipSPARSE <https://github.com/ROCm/hipSPARSE>`_
+    * - `hipSPARSE <https://github.com/ROCm/hipSPARSE>`__
       - :version-ref:`hipSPARSE rocm_version`
       - Accelerates operations on sparse matrices, such as sparse matrix-vector
         or matrix-matrix products.
       - Sparse tensor operations ``torch.sparse``.
-    * - `hipSPARSELt <https://github.com/ROCm/hipSPARSELt>`_
+    * - `hipSPARSELt <https://github.com/ROCm/hipSPARSELt>`__
       - :version-ref:`hipSPARSELt rocm_version`
       - Accelerates operations on sparse matrices, such as sparse matrix-vector
         or matrix-matrix products.
       - Sparse tensor operations ``torch.sparse``.
-    * - `hipTensor <https://github.com/ROCm/hipTensor>`_
+    * - `hipTensor <https://github.com/ROCm/hipTensor>`__
       - :version-ref:`hipTensor rocm_version`
       - Optimizes for high-performance tensor operations, such as contractions.
       - Accelerates tensor algebra, especially in deep learning and scientific
         computing.
-    * - `MIOpen <https://github.com/ROCm/MIOpen>`_
+    * - `MIOpen <https://github.com/ROCm/MIOpen>`__
       - :version-ref:`MIOpen rocm_version`
       - Optimizes deep learning primitives such as convolutions, pooling,
         normalization, and activation functions.
       - Speeds up convolutional neural networks (CNNs), recurrent neural
         networks (RNNs), and other layers. Used in operations like
         ``torch.nn.Conv2d``, ``torch.nn.ReLU``, and ``torch.nn.LSTM``.
-    * - `MIGraphX <https://github.com/ROCm/AMDMIGraphX>`_
+    * - `MIGraphX <https://github.com/ROCm/AMDMIGraphX>`__
       - :version-ref:`MIGraphX rocm_version`
       - Adds graph-level optimizations, ONNX models and mixed precision support
         and enable Ahead-of-Time (AOT) Compilation.
       - Speeds up inference models and executes ONNX models for
         compatibility with other frameworks.
         ``torch.nn.Conv2d``, ``torch.nn.ReLU``, and ``torch.nn.LSTM``.
-    * - `MIVisionX <https://github.com/ROCm/MIVisionX>`_
+    * - `MIVisionX <https://github.com/ROCm/MIVisionX>`__
       - :version-ref:`MIVisionX rocm_version`
       - Optimizes acceleration for computer vision and AI workloads like
         preprocessing, augmentation, and inferencing.
       - Faster data preprocessing and augmentation pipelines for datasets like
         ImageNet or COCO and easy to integrate into PyTorch's ``torch.utils.data``
         and ``torchvision`` workflows.
-    * - `rocAL <https://github.com/ROCm/rocAL>`_
+    * - `rocAL <https://github.com/ROCm/rocAL>`__
       - :version-ref:`rocAL rocm_version`
       - Accelerates the data pipeline by offloading intensive preprocessing and
         augmentation tasks. rocAL is part of MIVisionX.
       - Easy to integrate into PyTorch's ``torch.utils.data`` and
         ``torchvision`` data load workloads.
-    * - `RCCL <https://github.com/ROCm/rccl>`_
+    * - `RCCL <https://github.com/ROCm/rccl>`__
       - :version-ref:`RCCL rocm_version`
       - Optimizes for multi-GPU communication for operations like AllReduce and
         Broadcast.
       - Distributed data parallel training (``torch.nn.parallel.DistributedDataParallel``).
         Handles communication in multi-GPU setups.
-    * - `rocDecode <https://github.com/ROCm/rocDecode>`_
+    * - `rocDecode <https://github.com/ROCm/rocDecode>`__
       - :version-ref:`rocDecode rocm_version`
       - Provides hardware-accelerated data decoding capabilities, particularly
         for image, video, and other dataset formats.
       - Can be integrated in ``torch.utils.data``, ``torchvision.transforms``
         and ``torch.distributed``.
-    * - `rocJPEG <https://github.com/ROCm/rocJPEG>`_
+    * - `rocJPEG <https://github.com/ROCm/rocJPEG>`__
       - :version-ref:`rocJPEG rocm_version`
       - Provides hardware-accelerated JPEG image decoding and encoding.
       - GPU accelerated ``torchvision.io.decode_jpeg`` and
         ``torchvision.io.encode_jpeg`` and can be integrated in
         ``torch.utils.data`` and ``torchvision``.
-    * - `RPP <https://github.com/ROCm/RPP>`_
+    * - `RPP <https://github.com/ROCm/RPP>`__
       - :version-ref:`RPP rocm_version`
       - Speeds up data augmentation, transformation, and other preprocessing steps.
       - Easy to integrate into PyTorch's ``torch.utils.data`` and
         ``torchvision`` data load workloads to speed up data processing.
-    * - `rocThrust <https://github.com/ROCm/rocThrust>`_
+    * - `rocThrust <https://github.com/ROCm/rocThrust>`__
       - :version-ref:`rocThrust rocm_version`
       - Provides a C++ template library for parallel algorithms like sorting,
         reduction, and scanning.
       - Utilized in backend operations for tensor computations requiring
         parallel processing.
-    * - `rocWMMA <https://github.com/ROCm/rocWMMA>`_
+    * - `rocWMMA <https://github.com/ROCm/rocWMMA>`__
       - :version-ref:`rocWMMA rocm_version`
       - Accelerates warp-level matrix-multiply and matrix-accumulate to speed up matrix
         multiplication (GEMM) and accumulation operations with mixed precision
@@ -383,7 +383,7 @@ Supported data types
 The tensor data type is specified using the ``dtype`` attribute or argument.
 PyTorch supports many data types for different use cases.
 
-The following table lists `torch.Tensor <https://pytorch.org/docs/stable/tensors.html>`_
+The following table lists `torch.Tensor <https://pytorch.org/docs/stable/tensors.html>`__
 single data types:
 
 .. list-table::

docs/compatibility/ml-compatibility/tensorflow-compatibility.rst

@@ -10,16 +10,16 @@
 TensorFlow compatibility
 *******************************************************************************
 
-`TensorFlow <https://www.tensorflow.org/>`_ is an open-source library for
+`TensorFlow <https://www.tensorflow.org/>`__ is an open-source library for
 solving machine learning, deep learning, and AI problems. It can solve many
 problems across different sectors and industries but primarily focuses on
 neural network training and inference. It is one of the most popular and
 in-demand frameworks and is very active in open-source contribution and
 development.
 
-The `official TensorFlow repository <http://github.com/tensorflow/tensorflow>`_
+The `official TensorFlow repository <http://github.com/tensorflow/tensorflow>`__
 includes full ROCm support. AMD maintains a TensorFlow `ROCm repository
-<http://github.com/rocm/tensorflow-upstream>`_ in order to quickly add bug
+<http://github.com/rocm/tensorflow-upstream>`__ in order to quickly add bug
 fixes, updates, and support for the latest ROCM versions.
 
 - ROCm TensorFlow release:
@@ -27,16 +27,16 @@ fixes, updates, and support for the latest ROCM versions.
   - Offers :ref:`Docker images <tensorflow-docker-compat>` with
     ROCm and TensorFlow pre-installed.
 
-  - ROCm TensorFlow repository: `<https://github.com/ROCm/tensorflow-upstream>`_
+  - ROCm TensorFlow repository: `<https://github.com/ROCm/tensorflow-upstream>`__
 
   - See the :doc:`ROCm TensorFlow installation guide <rocm-install-on-linux:install/3rd-party/tensorflow-install>`
     to get started.
 
 - Official TensorFlow release:
 
-  - Official TensorFlow repository: `<https://github.com/tensorflow/tensorflow>`_
+  - Official TensorFlow repository: `<https://github.com/tensorflow/tensorflow>`__
 
-  - See the `TensorFlow API versions <https://www.tensorflow.org/versions>`_ list.
+  - See the `TensorFlow API versions <https://www.tensorflow.org/versions>`__ list.
 
   .. note::
 
@@ -54,9 +54,9 @@ Docker image compatibility
    <i class="fab fa-docker"></i>
 
 AMD validates and publishes ready-made `TensorFlow images
-<https://hub.docker.com/r/rocm/tensorflow>`_ with ROCm backends on
+<https://hub.docker.com/r/rocm/tensorflow>`__ with ROCm backends on
 Docker Hub. The following Docker image tags and associated inventories are
-validated for `ROCm 6.4.1 <https://repo.radeon.com/rocm/apt/6.4.1/>`_. Click
+validated for `ROCm 6.4.1 <https://repo.radeon.com/rocm/apt/6.4.1/>`__. Click
 the |docker-icon| icon to view the image on Docker Hub.
 
 .. list-table:: TensorFlow Docker image components
@@ -76,8 +76,8 @@ the |docker-icon| icon to view the image on Docker Hub.
       - `tensorflow-rocm 2.18.1 <https://repo.radeon.com/rocm/manylinux/rocm-rel-6.4.1/tensorflow_rocm-2.18.1-cp312-cp312-manylinux_2_28_x86_64.whl>`__
       - dev
       - 24.04
-      - `Python 3.12.10 <https://www.python.org/downloads/release/python-31210/>`_
+      - `Python 3.12.10 <https://www.python.org/downloads/release/python-31210/>`__
-      - `TensorBoard 2.18.0 <https://github.com/tensorflow/tensorboard/tree/2.18.0>`_
+      - `TensorBoard 2.18.0 <https://github.com/tensorflow/tensorboard/tree/2.18.0>`__
 
     * - .. raw:: html
 
@@ -86,8 +86,8 @@ the |docker-icon| icon to view the image on Docker Hub.
       - `tensorflow-rocm 2.18.1 <https://repo.radeon.com/rocm/manylinux/rocm-rel-6.4.1/tensorflow_rocm-2.18.1-cp312-cp312-manylinux_2_28_x86_64.whl>`__
       - runtime
       - 24.04
-      - `Python 3.12.10 <https://www.python.org/downloads/release/python-31210/>`_
+      - `Python 3.12.10 <https://www.python.org/downloads/release/python-31210/>`__
-      - `TensorBoard 2.18.0 <https://github.com/tensorflow/tensorboard/tree/2.18.0>`_
+      - `TensorBoard 2.18.0 <https://github.com/tensorflow/tensorboard/tree/2.18.0>`__
 
     * - .. raw:: html
 
@@ -96,8 +96,8 @@ the |docker-icon| icon to view the image on Docker Hub.
       - `tensorflow-rocm 2.18.1 <https://repo.radeon.com/rocm/manylinux/rocm-rel-6.4.1/tensorflow_rocm-2.18.1-cp310-cp310-manylinux_2_28_x86_64.whl>`__
       - dev
       - 22.04
-      - `Python 3.10.17 <https://www.python.org/downloads/release/python-31017/>`_
+      - `Python 3.10.17 <https://www.python.org/downloads/release/python-31017/>`__
-      - `TensorBoard 2.18.0 <https://github.com/tensorflow/tensorboard/tree/2.18.0>`_
+      - `TensorBoard 2.18.0 <https://github.com/tensorflow/tensorboard/tree/2.18.0>`__
 
     * - .. raw:: html
 
@@ -106,8 +106,8 @@ the |docker-icon| icon to view the image on Docker Hub.
       - `tensorflow-rocm 2.18.1 <https://repo.radeon.com/rocm/manylinux/rocm-rel-6.4.1/tensorflow_rocm-2.18.1-cp310-cp310-manylinux_2_28_x86_64.whl>`__
       - runtime
       - 22.04
-      - `Python 3.10.17 <https://www.python.org/downloads/release/python-31017/>`_
+      - `Python 3.10.17 <https://www.python.org/downloads/release/python-31017/>`__
-      - `TensorBoard 2.18.0 <https://github.com/tensorflow/tensorboard/tree/2.18.0>`_
+      - `TensorBoard 2.18.0 <https://github.com/tensorflow/tensorboard/tree/2.18.0>`__
 
     * - .. raw:: html
 
@@ -116,8 +116,8 @@ the |docker-icon| icon to view the image on Docker Hub.
       - `tensorflow-rocm 2.17.1 <https://repo.radeon.com/rocm/manylinux/rocm-rel-6.4/tensorflow_rocm-2.17.1-cp312-cp312-manylinux_2_28_x86_64.whl>`__
       - dev
       - 24.04
-      - `Python 3.12.10 <https://www.python.org/downloads/release/python-31210/>`_
+      - `Python 3.12.10 <https://www.python.org/downloads/release/python-31210/>`__
-      - `TensorBoard 2.17.1 <https://github.com/tensorflow/tensorboard/tree/2.17.1>`_
+      - `TensorBoard 2.17.1 <https://github.com/tensorflow/tensorboard/tree/2.17.1>`__
 
     * - .. raw:: html
 
@@ -126,8 +126,8 @@ the |docker-icon| icon to view the image on Docker Hub.
       - `tensorflow-rocm 2.18.1 <https://repo.radeon.com/rocm/manylinux/rocm-rel-6.4/tensorflow_rocm-2.17.1-cp312-cp312-manylinux_2_28_x86_64.whl>`__
       - runtime
       - 24.04
-      - `Python 3.12.10 <https://www.python.org/downloads/release/python-3124/>`_
+      - `Python 3.12.10 <https://www.python.org/downloads/release/python-3124/>`__
-      - `TensorBoard 2.17.1 <https://github.com/tensorflow/tensorboard/tree/2.17.1>`_
+      - `TensorBoard 2.17.1 <https://github.com/tensorflow/tensorboard/tree/2.17.1>`__
 
     * - .. raw:: html
 
@@ -136,8 +136,8 @@ the |docker-icon| icon to view the image on Docker Hub.
       - `tensorflow-rocm 2.17.1 <https://repo.radeon.com/rocm/manylinux/rocm-rel-6.4.1/tensorflow_rocm-2.17.1-cp312-cp312-manylinux_2_28_x86_64.whl>`__
       - dev
       - 22.04
-      - `Python 3.10.17 <https://www.python.org/downloads/release/python-31017/>`_
+      - `Python 3.10.17 <https://www.python.org/downloads/release/python-31017/>`__
-      - `TensorBoard 2.17.1 <https://github.com/tensorflow/tensorboard/tree/2.17.1>`_
+      - `TensorBoard 2.17.1 <https://github.com/tensorflow/tensorboard/tree/2.17.1>`__
 
     * - .. raw:: html
 
@@ -146,8 +146,8 @@ the |docker-icon| icon to view the image on Docker Hub.
       - `tensorflow-rocm 2.17.1 <https://repo.radeon.com/rocm/manylinux/rocm-rel-6.4.1/tensorflow_rocm-2.17.1-cp310-cp310-manylinux_2_28_x86_64.whl>`__
       - runtime
       - 22.04
-      - `Python 3.10.17 <https://www.python.org/downloads/release/python-31017/>`_
+      - `Python 3.10.17 <https://www.python.org/downloads/release/python-31017/>`__
-      - `TensorBoard 2.17.1 <https://github.com/tensorflow/tensorboard/tree/2.17.1>`_
+      - `TensorBoard 2.17.1 <https://github.com/tensorflow/tensorboard/tree/2.17.1>`__
 
     * - .. raw:: html
 
@@ -156,8 +156,8 @@ the |docker-icon| icon to view the image on Docker Hub.
       - `tensorflow-rocm 2.16.2 <https://repo.radeon.com/rocm/manylinux/rocm-rel-6.4.1/tensorflow_rocm-2.16.2-cp312-cp312-manylinux_2_28_x86_64.whl>`__
       - dev
       - 24.04
-      - `Python 3.12.10 <https://www.python.org/downloads/release/python-31210/>`_
+      - `Python 3.12.10 <https://www.python.org/downloads/release/python-31210/>`__
-      - `TensorBoard 2.16.2 <https://github.com/tensorflow/tensorboard/tree/2.16.2>`_
+      - `TensorBoard 2.16.2 <https://github.com/tensorflow/tensorboard/tree/2.16.2>`__
 
     * - .. raw:: html
 
@@ -166,8 +166,8 @@ the |docker-icon| icon to view the image on Docker Hub.
       - `tensorflow-rocm 2.16.2 <https://repo.radeon.com/rocm/manylinux/rocm-rel-6.4.1/tensorflow_rocm-2.16.2-cp312-cp312-manylinux_2_28_x86_64.whl>`__
       - runtime
       - 24.04
-      - `Python 3.12.10 <https://www.python.org/downloads/release/python-31210/>`_
+      - `Python 3.12.10 <https://www.python.org/downloads/release/python-31210/>`__
-      - `TensorBoard 2.16.2 <https://github.com/tensorflow/tensorboard/tree/2.16.2>`_
+      - `TensorBoard 2.16.2 <https://github.com/tensorflow/tensorboard/tree/2.16.2>`__
 
     * - .. raw:: html
 
@@ -176,8 +176,8 @@ the |docker-icon| icon to view the image on Docker Hub.
       - `tensorflow-rocm 2.16.2 <https://hub.docker.com/layers/rocm/tensorflow/rocm6.4.1-py3.10-tf2.16-dev/images/sha256-36c4fa047c86e2470ac473ec1429aea6d4b8934b90ffeb34d1afab40e7e5b377>`__
       - dev
       - 22.04
-      - `Python 3.10.17 <https://www.python.org/downloads/release/python-31017/>`_
+      - `Python 3.10.17 <https://www.python.org/downloads/release/python-31017/>`__
-      - `TensorBoard 2.16.2 <https://github.com/tensorflow/tensorboard/tree/2.16.2>`_
+      - `TensorBoard 2.16.2 <https://github.com/tensorflow/tensorboard/tree/2.16.2>`__
 
     * - .. raw:: html
 
@@ -186,8 +186,8 @@ the |docker-icon| icon to view the image on Docker Hub.
       - `tensorflow-rocm 2.16.2 <https://repo.radeon.com/rocm/manylinux/rocm-rel-6.4.1/tensorflow_rocm-2.16.2-cp312-cp312-manylinux_2_28_x86_64.whl>`__
       - runtime
       - 22.04
-      - `Python 3.10.17 <https://www.python.org/downloads/release/python-31017/>`_
+      - `Python 3.10.17 <https://www.python.org/downloads/release/python-31017/>`__
-      - `TensorBoard 2.16.2 <https://github.com/tensorflow/tensorboard/tree/2.16.2>`_
+      - `TensorBoard 2.16.2 <https://github.com/tensorflow/tensorboard/tree/2.16.2>`__
 
 
 Critical ROCm libraries for TensorFlow
@@ -207,43 +207,43 @@ are available in ROCm :version:`rocm_version`.
       - Version
       - Purpose
       - Used in
-    * - `hipBLAS <https://github.com/ROCm/hipBLAS>`_
+    * - `hipBLAS <https://github.com/ROCm/hipBLAS>`__
       - :version-ref:`hipBLAS rocm_version`
       - Provides GPU-accelerated Basic Linear Algebra Subprograms (BLAS) for
         matrix and vector operations.
       - Accelerates operations like ``tf.matmul``, ``tf.linalg.matmul``, and
         other matrix multiplications commonly used in neural network layers.
-    * - `hipBLASLt <https://github.com/ROCm/hipBLASLt>`_
+    * - `hipBLASLt <https://github.com/ROCm/hipBLASLt>`__
       - :version-ref:`hipBLASLt rocm_version`
       - Extends hipBLAS with additional optimizations like fused kernels and
         integer tensor cores.
       - Optimizes matrix multiplications and linear algebra operations used in
         layers like dense, convolutional, and RNNs in TensorFlow.
-    * - `hipCUB <https://github.com/ROCm/hipCUB>`_
+    * - `hipCUB <https://github.com/ROCm/hipCUB>`__
       - :version-ref:`hipCUB rocm_version`
       - Provides a C++ template library for parallel algorithms for reduction,
         scan, sort and select.
       - Supports operations like ``tf.reduce_sum``, ``tf.cumsum``, ``tf.sort``
         and other tensor operations in TensorFlow, especially those involving
         scanning, sorting, and filtering.
-    * - `hipFFT <https://github.com/ROCm/hipFFT>`_
+    * - `hipFFT <https://github.com/ROCm/hipFFT>`__
       - :version-ref:`hipFFT rocm_version`
       - Accelerates Fast Fourier Transforms (FFT) for signal processing tasks.
       - Used for operations like signal processing, image filtering, and
         certain types of neural networks requiring FFT-based transformations.
-    * - `hipSOLVER <https://github.com/ROCm/hipSOLVER>`_
+    * - `hipSOLVER <https://github.com/ROCm/hipSOLVER>`__
       - :version-ref:`hipSOLVER rocm_version`
       - Provides GPU-accelerated direct linear solvers for dense and sparse
         systems.
       - Optimizes linear algebra functions such as solving systems of linear
         equations, often used in optimization and training tasks.
-    * - `hipSPARSE <https://github.com/ROCm/hipSPARSE>`_
+    * - `hipSPARSE <https://github.com/ROCm/hipSPARSE>`__
       - :version-ref:`hipSPARSE rocm_version`
       - Optimizes sparse matrix operations for efficient computations on sparse
         data.
       - Accelerates sparse matrix operations in models with sparse weight
         matrices or activations, commonly used in neural networks.
-    * - `MIOpen <https://github.com/ROCm/MIOpen>`_
+    * - `MIOpen <https://github.com/ROCm/MIOpen>`__
       - :version-ref:`MIOpen rocm_version`
       - Provides optimized deep learning primitives such as convolutions,
         pooling,
@@ -251,13 +251,13 @@ are available in ROCm :version:`rocm_version`.
       - Speeds up convolutional neural networks (CNNs) and other layers. Used
         in TensorFlow for layers like ``tf.nn.conv2d``, ``tf.nn.relu``, and
         ``tf.nn.lstm_cell``.
-    * - `RCCL <https://github.com/ROCm/rccl>`_
+    * - `RCCL <https://github.com/ROCm/rccl>`__
       - :version-ref:`RCCL rocm_version`
       - Optimizes for multi-GPU communication for operations like AllReduce and
         Broadcast.
       - Distributed data parallel training (``tf.distribute.MirroredStrategy``).
         Handles communication in multi-GPU setups.
-    * - `rocThrust <https://github.com/ROCm/rocThrust>`_
+    * - `rocThrust <https://github.com/ROCm/rocThrust>`__
       - :version-ref:`rocThrust rocm_version`
       - Provides a C++ template library for parallel algorithms like sorting,
         reduction, and scanning.
@@ -278,7 +278,7 @@ The data type of a tensor is specified using the ``dtype`` attribute or
 argument, and TensorFlow supports a wide range of data types for different use
 cases.
 
-The basic, single data types of `tf.dtypes <https://www.tensorflow.org/api_docs/python/tf/dtypes>`_
+The basic, single data types of `tf.dtypes <https://www.tensorflow.org/api_docs/python/tf/dtypes>`__
 are as follows:
 
 .. list-table::
@@ -550,7 +550,7 @@ Use cases and recommendations
 ===============================================================================
 
 * The `Training a Neural Collaborative Filtering (NCF) Recommender on an AMD
-  GPU <https://rocm.blogs.amd.com/artificial-intelligence/ncf/README.html>`_
+  GPU <https://rocm.blogs.amd.com/artificial-intelligence/ncf/README.html>`__
   blog post discusses training an NCF recommender system using TensorFlow. It
   explains how NCF improves traditional collaborative filtering methods by
   leveraging neural networks to model non-linear user-item interactions. The
@@ -559,7 +559,7 @@ Use cases and recommendations
   purchasing) and how it addresses challenges like the lack of negative values.
 
 * The `Creating a PyTorch/TensorFlow code environment on AMD GPUs
-  <https://rocm.blogs.amd.com/software-tools-optimization/pytorch-tensorflow-env/README.html>`_
+  <https://rocm.blogs.amd.com/software-tools-optimization/pytorch-tensorflow-env/README.html>`__
   blog post provides instructions for creating a machine learning environment
   for PyTorch and TensorFlow on AMD GPUs using ROCm. It covers steps like
   installing the libraries, cloning code repositories, installing dependencies,
@@ -568,4 +568,4 @@ Use cases and recommendations
   for a better experience on AMD GPUs. This guide aims to help data scientists
   and ML practitioners adapt their code for AMD GPUs.
 
-For more use cases and recommendations, see the `ROCm Tensorflow blog posts <https://rocm.blogs.amd.com/blog/tag/tensorflow.html>`_.
+For more use cases and recommendations, see the `ROCm Tensorflow blog posts <https://rocm.blogs.amd.com/blog/tag/tensorflow.html>`__.

docs/compatibility/ml-compatibility/verl-compatibility.rst

@@ -16,7 +16,7 @@ verl offers a scalable, open-source fine-tuning solution optimized for AMD Insti
 * See the `verl documentation <https://verl.readthedocs.io/en/latest/>`_ for more information about verl. 
 * The official verl GitHub repository is `https://github.com/volcengine/verl <https://github.com/volcengine/verl>`_.
 * Use the AMD-validated :ref:`Docker images <verl-docker-compat>` with ROCm and verl preinstalled. 
-* See the :doc:`ROCm verl installation guide <rocm-install-on-linux:install/3rd-party/dgl-install>` to get started.
+* See the :doc:`ROCm verl installation guide <rocm-install-on-linux:install/3rd-party/verl-install>` to get started.
 
 .. note::
 

docs/conceptual/gpu-arch/mi300-mi200-performance-counters.rst

@@ -8,7 +8,7 @@ MI300 and MI200 series performance counters and metrics
 
 This document lists and describes the hardware performance counters and derived metrics available
 for the AMD Instinct™ MI300 and MI200 GPU. You can also access this information using the
-:doc:`ROCProfiler tool <rocprofiler:rocprofv1>`.
+:doc:`ROCprofiler-SDK <rocprofiler-sdk:how-to/using-rocprofv3>`.
 
 MI300 and MI200 series performance counters
 ===============================================================

docs/conf.py

@@ -12,6 +12,54 @@ from pathlib import Path
 shutil.copy2("../RELEASE.md", "./about/release-notes.md")
 shutil.copy2("../CHANGELOG.md", "./release/changelog.md")
 
+# Mark the consolidated changelog as orphan to prevent Sphinx from warning about missing toctree entries
+with open("./release/changelog.md", "r+") as file:
+    content = file.read()
+    file.seek(0)
+    file.write(":orphan:\n" + content)
+
+# Replace GitHub-style [!ADMONITION]s with Sphinx-compatible ```{admonition} blocks
+with open("./release/changelog.md", "r") as file:
+    lines = file.readlines()
+
+    modified_lines = []
+    in_admonition_section = False
+
+    # Map for matching the specific admonition type to its corresponding Sphinx markdown syntax
+    admonition_types = {
+        '> [!NOTE]': '```{note}',
+        '> [!TIP]': '```{tip}',
+        '> [!IMPORTANT]': '```{important}',
+        '> [!WARNING]': '```{warning}',
+        '> [!CAUTION]': '```{caution}'
+    }
+
+    for line in lines:
+        if any(line.startswith(k) for k in admonition_types):
+            for key in admonition_types:
+                if(line.startswith(key)):
+                    modified_lines.append(admonition_types[key] + '\n')
+                    break
+            in_admonition_section = True
+        elif in_admonition_section:
+            if line.strip() == '':
+                # If we encounter an empty line, close the admonition section
+                modified_lines.append('```\n\n')  # Close the admonition block
+                in_admonition_section = False
+            else:
+                modified_lines.append(line.lstrip('> '))
+        else:
+            modified_lines.append(line)
+
+    # In case the file ended while still in a admonition section, close it
+    if in_admonition_section:
+        modified_lines.append('```')
+
+    file.close()
+
+    with open("./release/changelog.md", 'w') as file:
+        file.writelines(modified_lines)
+
 os.system("mkdir -p ../_readthedocs/html/downloads")
 os.system("cp compatibility/compatibility-matrix-historical-6.0.csv ../_readthedocs/html/downloads/compatibility-matrix-historical-6.0.csv")
 
@@ -57,10 +105,22 @@ article_pages = [
     {"file": "how-to/rocm-for-ai/training/index", "os": ["linux"]},
     {"file": "how-to/rocm-for-ai/training/train-a-model", "os": ["linux"]},
     {"file": "how-to/rocm-for-ai/training/prerequisite-system-validation", "os": ["linux"]},
-    {"file": "how-to/rocm-for-ai/training/benchmark-docker/megatron-lm", "os": ["linux"]},
-    {"file": "how-to/rocm-for-ai/training/benchmark-docker/pytorch-training", "os": ["linux"]},
-    {"file": "how-to/rocm-for-ai/training/benchmark-docker/mpt-llm-foundry", "os": ["linux"]},
     {"file": "how-to/rocm-for-ai/training/scale-model-training", "os": ["linux"]},
+    {"file": "how-to/rocm-for-ai/training/benchmark-docker/megatron-lm", "os": ["linux"]},
+    {"file": "how-to/rocm-for-ai/training/benchmark-docker/previous-versions/megatron-lm-history", "os": ["linux"]},
+    {"file": "how-to/rocm-for-ai/training/benchmark-docker/previous-versions/megatron-lm-v24.12-dev", "os": ["linux"]},
+    {"file": "how-to/rocm-for-ai/training/benchmark-docker/previous-versions/megatron-lm-v25.3", "os": ["linux"]},
+    {"file": "how-to/rocm-for-ai/training/benchmark-docker/previous-versions/megatron-lm-v25.4", "os": ["linux"]},
+    {"file": "how-to/rocm-for-ai/training/benchmark-docker/previous-versions/megatron-lm-v25.5", "os": ["linux"]},
+    {"file": "how-to/rocm-for-ai/training/benchmark-docker/pytorch-training", "os": ["linux"]},
+    {"file": "how-to/rocm-for-ai/training/benchmark-docker/previous-versions/pytorch-training-history", "os": ["linux"]},
+    {"file": "how-to/rocm-for-ai/training/benchmark-docker/previous-versions/pytorch-training-v25.3", "os": ["linux"]},
+    {"file": "how-to/rocm-for-ai/training/benchmark-docker/previous-versions/pytorch-training-v25.4", "os": ["linux"]},
+    {"file": "how-to/rocm-for-ai/training/benchmark-docker/previous-versions/pytorch-training-v25.5", "os": ["linux"]},
+    {"file": "how-to/rocm-for-ai/training/benchmark-docker/jax-maxtext", "os": ["linux"]},
+    {"file": "how-to/rocm-for-ai/training/benchmark-docker/previous-versions/jax-maxtext-history", "os": ["linux"]},
+    {"file": "how-to/rocm-for-ai/training/benchmark-docker/previous-versions/jax-maxtext-v25.4", "os": ["linux"]},
+    {"file": "how-to/rocm-for-ai/training/benchmark-docker/mpt-llm-foundry", "os": ["linux"]},
 
     {"file": "how-to/rocm-for-ai/fine-tuning/index", "os": ["linux"]},
     {"file": "how-to/rocm-for-ai/fine-tuning/overview", "os": ["linux"]},
@@ -72,7 +132,16 @@ article_pages = [
     {"file": "how-to/rocm-for-ai/inference/hugging-face-models", "os": ["linux"]},
     {"file": "how-to/rocm-for-ai/inference/llm-inference-frameworks", "os": ["linux"]},
     {"file": "how-to/rocm-for-ai/inference/benchmark-docker/vllm", "os": ["linux"]},
+    {"file": "how-to/rocm-for-ai/inference/benchmark-docker/previous-versions/vllm-history", "os": ["linux"]},
+    {"file": "how-to/rocm-for-ai/inference/benchmark-docker/previous-versions/vllm-0.4.3", "os": ["linux"]},
+    {"file": "how-to/rocm-for-ai/inference/benchmark-docker/previous-versions/vllm-0.6.4", "os": ["linux"]},
+    {"file": "how-to/rocm-for-ai/inference/benchmark-docker/previous-versions/vllm-0.6.6", "os": ["linux"]},
+    {"file": "how-to/rocm-for-ai/inference/benchmark-docker/previous-versions/vllm-0.7.3-20250325", "os": ["linux"]},
+    {"file": "how-to/rocm-for-ai/inference/benchmark-docker/previous-versions/vllm-0.8.3-20250415", "os": ["linux"]},
     {"file": "how-to/rocm-for-ai/inference/benchmark-docker/previous-versions/vllm-0.8.5-20250513", "os": ["linux"]},
+    {"file": "how-to/rocm-for-ai/inference/benchmark-docker/previous-versions/vllm-0.8.5-20250521", "os": ["linux"]},
+    {"file": "how-to/rocm-for-ai/inference/benchmark-docker/previous-versions/vllm-0.9.0.1-20250605", "os": ["linux"]},
+    {"file": "how-to/rocm-for-ai/inference/benchmark-docker/previous-versions/vllm-0.9.0.1-20250702", "os": ["linux"]},
     {"file": "how-to/rocm-for-ai/inference/benchmark-docker/pytorch-inference", "os": ["linux"]},
     {"file": "how-to/rocm-for-ai/inference/deploy-your-model", "os": ["linux"]},
 
@@ -129,6 +198,7 @@ html_theme_options = {"link_main_doc": False}
 redirects = {"reference/openmp/openmp": "../../about/compatibility/openmp.html"}
 
 numfig = False
+suppress_warnings = ["autosectionlabel.*"]
 
 html_context = {
     "project_path" : {project_path},

docs/data/how-to/rocm-for-ai/inference/previous-versions/vllm_0.9.0.1_20250605-benchmark-models.yaml

@@ -0,0 +1,162 @@
+vllm_benchmark:
+  unified_docker:
+    latest:
+      pull_tag: rocm/vllm:rocm6.4.1_vllm_0.9.0.1_20250605
+      docker_hub_url: https://hub.docker.com/layers/rocm/vllm/rocm6.4.1_vllm_0.9.0.1_20250605/images/sha256-f48beeb3d72663a93c77211eb45273d564451447c097e060befa713d565fa36c
+      rocm_version: 6.4.1
+      vllm_version: 0.9.0.1 (0.9.0.2.dev108+g71faa1880.rocm641)
+      pytorch_version: 2.7.0+gitf717b2a
+      hipblaslt_version: 0.15
+  model_groups:
+    - group: Meta Llama
+      tag: llama
+      models:
+      - model: Llama 3.1 8B
+        mad_tag: pyt_vllm_llama-3.1-8b
+        model_repo: meta-llama/Llama-3.1-8B-Instruct
+        url: https://huggingface.co/meta-llama/Llama-3.1-8B
+        precision: float16
+      - model: Llama 3.1 70B
+        mad_tag: pyt_vllm_llama-3.1-70b
+        model_repo: meta-llama/Llama-3.1-70B-Instruct
+        url: https://huggingface.co/meta-llama/Llama-3.1-70B-Instruct
+        precision: float16
+      - model: Llama 3.1 405B
+        mad_tag: pyt_vllm_llama-3.1-405b
+        model_repo: meta-llama/Llama-3.1-405B-Instruct
+        url: https://huggingface.co/meta-llama/Llama-3.1-405B-Instruct
+        precision: float16
+      - model: Llama 2 7B
+        mad_tag: pyt_vllm_llama-2-7b
+        model_repo: meta-llama/Llama-2-7b-chat-hf
+        url: https://huggingface.co/meta-llama/Llama-2-7b-chat-hf
+        precision: float16
+      - model: Llama 2 70B
+        mad_tag: pyt_vllm_llama-2-70b
+        model_repo: meta-llama/Llama-2-70b-chat-hf
+        url: https://huggingface.co/meta-llama/Llama-2-70b-chat-hf
+        precision: float16
+      - model: Llama 3.1 8B FP8
+        mad_tag: pyt_vllm_llama-3.1-8b_fp8
+        model_repo: amd/Llama-3.1-8B-Instruct-FP8-KV
+        url: https://huggingface.co/amd/Llama-3.1-8B-Instruct-FP8-KV
+        precision: float8
+      - model: Llama 3.1 70B FP8
+        mad_tag: pyt_vllm_llama-3.1-70b_fp8
+        model_repo: amd/Llama-3.1-70B-Instruct-FP8-KV
+        url: https://huggingface.co/amd/Llama-3.1-70B-Instruct-FP8-KV
+        precision: float8
+      - model: Llama 3.1 405B FP8
+        mad_tag: pyt_vllm_llama-3.1-405b_fp8
+        model_repo: amd/Llama-3.1-405B-Instruct-FP8-KV
+        url: https://huggingface.co/amd/Llama-3.1-405B-Instruct-FP8-KV
+        precision: float8
+    - group: Mistral AI
+      tag: mistral
+      models:
+      - model: Mixtral MoE 8x7B
+        mad_tag: pyt_vllm_mixtral-8x7b
+        model_repo: mistralai/Mixtral-8x7B-Instruct-v0.1
+        url: https://huggingface.co/mistralai/Mixtral-8x7B-Instruct-v0.1
+        precision: float16
+      - model: Mixtral MoE 8x22B
+        mad_tag: pyt_vllm_mixtral-8x22b
+        model_repo: mistralai/Mixtral-8x22B-Instruct-v0.1
+        url: https://huggingface.co/mistralai/Mixtral-8x22B-Instruct-v0.1
+        precision: float16
+      - model: Mistral 7B
+        mad_tag: pyt_vllm_mistral-7b
+        model_repo: mistralai/Mistral-7B-Instruct-v0.3
+        url: https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.3
+        precision: float16
+      - model: Mixtral MoE 8x7B FP8
+        mad_tag: pyt_vllm_mixtral-8x7b_fp8
+        model_repo: amd/Mixtral-8x7B-Instruct-v0.1-FP8-KV
+        url: https://huggingface.co/amd/Mixtral-8x7B-Instruct-v0.1-FP8-KV
+        precision: float8
+      - model: Mixtral MoE 8x22B FP8
+        mad_tag: pyt_vllm_mixtral-8x22b_fp8
+        model_repo: amd/Mixtral-8x22B-Instruct-v0.1-FP8-KV
+        url: https://huggingface.co/amd/Mixtral-8x22B-Instruct-v0.1-FP8-KV
+        precision: float8
+      - model: Mistral 7B FP8
+        mad_tag: pyt_vllm_mistral-7b_fp8
+        model_repo: amd/Mistral-7B-v0.1-FP8-KV
+        url: https://huggingface.co/amd/Mistral-7B-v0.1-FP8-KV
+        precision: float8
+    - group: Qwen
+      tag: qwen
+      models:
+      - model: Qwen2 7B
+        mad_tag: pyt_vllm_qwen2-7b
+        model_repo: Qwen/Qwen2-7B-Instruct
+        url: https://huggingface.co/Qwen/Qwen2-7B-Instruct
+        precision: float16
+      - model: Qwen2 72B
+        mad_tag: pyt_vllm_qwen2-72b
+        model_repo: Qwen/Qwen2-72B-Instruct
+        url: https://huggingface.co/Qwen/Qwen2-72B-Instruct
+        precision: float16
+      - model: QwQ-32B
+        mad_tag: pyt_vllm_qwq-32b
+        model_repo: Qwen/QwQ-32B
+        url: https://huggingface.co/Qwen/QwQ-32B
+        precision: float16
+        tunableop: true
+    - group: Databricks DBRX
+      tag: dbrx
+      models:
+      - model: DBRX Instruct
+        mad_tag: pyt_vllm_dbrx-instruct
+        model_repo: databricks/dbrx-instruct
+        url: https://huggingface.co/databricks/dbrx-instruct
+        precision: float16
+      - model: DBRX Instruct FP8
+        mad_tag: pyt_vllm_dbrx_fp8
+        model_repo: amd/dbrx-instruct-FP8-KV
+        url: https://huggingface.co/amd/dbrx-instruct-FP8-KV
+        precision: float8
+    - group: Google Gemma
+      tag: gemma
+      models:
+      - model: Gemma 2 27B
+        mad_tag: pyt_vllm_gemma-2-27b
+        model_repo: google/gemma-2-27b
+        url: https://huggingface.co/google/gemma-2-27b
+        precision: float16
+    - group: Cohere
+      tag: cohere
+      models:
+      - model: C4AI Command R+ 08-2024
+        mad_tag: pyt_vllm_c4ai-command-r-plus-08-2024
+        model_repo: CohereForAI/c4ai-command-r-plus-08-2024
+        url: https://huggingface.co/CohereForAI/c4ai-command-r-plus-08-2024
+        precision: float16
+      - model: C4AI Command R+ 08-2024 FP8
+        mad_tag: pyt_vllm_command-r-plus_fp8
+        model_repo: amd/c4ai-command-r-plus-FP8-KV
+        url: https://huggingface.co/amd/c4ai-command-r-plus-FP8-KV
+        precision: float8
+    - group: DeepSeek
+      tag: deepseek
+      models:
+      - model: DeepSeek MoE 16B
+        mad_tag: pyt_vllm_deepseek-moe-16b-chat
+        model_repo: deepseek-ai/deepseek-moe-16b-chat
+        url: https://huggingface.co/deepseek-ai/deepseek-moe-16b-chat
+        precision: float16
+    - group: Microsoft Phi
+      tag: phi
+      models:
+      - model: Phi-4
+        mad_tag: pyt_vllm_phi-4
+        model_repo: microsoft/phi-4
+        url: https://huggingface.co/microsoft/phi-4
+    - group: TII Falcon
+      tag: falcon
+      models:
+      - model: Falcon 180B
+        mad_tag: pyt_vllm_falcon-180b
+        model_repo: tiiuae/falcon-180B
+        url: https://huggingface.co/tiiuae/falcon-180B
+        precision: float16

docs/data/how-to/rocm-for-ai/inference/previous-versions/vllm_0.9.1_20250702-benchmark-models.yaml

@@ -0,0 +1,163 @@
+vllm_benchmark:
+  unified_docker:
+    latest:
+      # TODO: update me
+      pull_tag: rocm/vllm:rocm6.4.1_vllm_0.9.1_20250702
+      docker_hub_url: https://hub.docker.com/layers/rocm/vllm/rocm6.4.1_vllm_0.9.1_20250702/images/sha256-45068a2079cb8df554ed777141bf0c67d6627c470a897256e60c9f262677faab
+      rocm_version: 6.4.1
+      vllm_version: 0.9.1 (0.9.2.dev206+gb335519f2.rocm641)
+      pytorch_version: 2.7.0+gitf717b2a
+      hipblaslt_version: 0.15
+  model_groups:
+    - group: Meta Llama
+      tag: llama
+      models:
+      - model: Llama 3.1 8B
+        mad_tag: pyt_vllm_llama-3.1-8b
+        model_repo: meta-llama/Llama-3.1-8B-Instruct
+        url: https://huggingface.co/meta-llama/Llama-3.1-8B
+        precision: float16
+      - model: Llama 3.1 70B
+        mad_tag: pyt_vllm_llama-3.1-70b
+        model_repo: meta-llama/Llama-3.1-70B-Instruct
+        url: https://huggingface.co/meta-llama/Llama-3.1-70B-Instruct
+        precision: float16
+      - model: Llama 3.1 405B
+        mad_tag: pyt_vllm_llama-3.1-405b
+        model_repo: meta-llama/Llama-3.1-405B-Instruct
+        url: https://huggingface.co/meta-llama/Llama-3.1-405B-Instruct
+        precision: float16
+      - model: Llama 2 7B
+        mad_tag: pyt_vllm_llama-2-7b
+        model_repo: meta-llama/Llama-2-7b-chat-hf
+        url: https://huggingface.co/meta-llama/Llama-2-7b-chat-hf
+        precision: float16
+      - model: Llama 2 70B
+        mad_tag: pyt_vllm_llama-2-70b
+        model_repo: meta-llama/Llama-2-70b-chat-hf
+        url: https://huggingface.co/meta-llama/Llama-2-70b-chat-hf
+        precision: float16
+      - model: Llama 3.1 8B FP8
+        mad_tag: pyt_vllm_llama-3.1-8b_fp8
+        model_repo: amd/Llama-3.1-8B-Instruct-FP8-KV
+        url: https://huggingface.co/amd/Llama-3.1-8B-Instruct-FP8-KV
+        precision: float8
+      - model: Llama 3.1 70B FP8
+        mad_tag: pyt_vllm_llama-3.1-70b_fp8
+        model_repo: amd/Llama-3.1-70B-Instruct-FP8-KV
+        url: https://huggingface.co/amd/Llama-3.1-70B-Instruct-FP8-KV
+        precision: float8
+      - model: Llama 3.1 405B FP8
+        mad_tag: pyt_vllm_llama-3.1-405b_fp8
+        model_repo: amd/Llama-3.1-405B-Instruct-FP8-KV
+        url: https://huggingface.co/amd/Llama-3.1-405B-Instruct-FP8-KV
+        precision: float8
+    - group: Mistral AI
+      tag: mistral
+      models:
+      - model: Mixtral MoE 8x7B
+        mad_tag: pyt_vllm_mixtral-8x7b
+        model_repo: mistralai/Mixtral-8x7B-Instruct-v0.1
+        url: https://huggingface.co/mistralai/Mixtral-8x7B-Instruct-v0.1
+        precision: float16
+      - model: Mixtral MoE 8x22B
+        mad_tag: pyt_vllm_mixtral-8x22b
+        model_repo: mistralai/Mixtral-8x22B-Instruct-v0.1
+        url: https://huggingface.co/mistralai/Mixtral-8x22B-Instruct-v0.1
+        precision: float16
+      - model: Mistral 7B
+        mad_tag: pyt_vllm_mistral-7b
+        model_repo: mistralai/Mistral-7B-Instruct-v0.3
+        url: https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.3
+        precision: float16
+      - model: Mixtral MoE 8x7B FP8
+        mad_tag: pyt_vllm_mixtral-8x7b_fp8
+        model_repo: amd/Mixtral-8x7B-Instruct-v0.1-FP8-KV
+        url: https://huggingface.co/amd/Mixtral-8x7B-Instruct-v0.1-FP8-KV
+        precision: float8
+      - model: Mixtral MoE 8x22B FP8
+        mad_tag: pyt_vllm_mixtral-8x22b_fp8
+        model_repo: amd/Mixtral-8x22B-Instruct-v0.1-FP8-KV
+        url: https://huggingface.co/amd/Mixtral-8x22B-Instruct-v0.1-FP8-KV
+        precision: float8
+      - model: Mistral 7B FP8
+        mad_tag: pyt_vllm_mistral-7b_fp8
+        model_repo: amd/Mistral-7B-v0.1-FP8-KV
+        url: https://huggingface.co/amd/Mistral-7B-v0.1-FP8-KV
+        precision: float8
+    - group: Qwen
+      tag: qwen
+      models:
+      - model: Qwen2 7B
+        mad_tag: pyt_vllm_qwen2-7b
+        model_repo: Qwen/Qwen2-7B-Instruct
+        url: https://huggingface.co/Qwen/Qwen2-7B-Instruct
+        precision: float16
+      - model: Qwen2 72B
+        mad_tag: pyt_vllm_qwen2-72b
+        model_repo: Qwen/Qwen2-72B-Instruct
+        url: https://huggingface.co/Qwen/Qwen2-72B-Instruct
+        precision: float16
+      - model: QwQ-32B
+        mad_tag: pyt_vllm_qwq-32b
+        model_repo: Qwen/QwQ-32B
+        url: https://huggingface.co/Qwen/QwQ-32B
+        precision: float16
+        tunableop: true
+    - group: Databricks DBRX
+      tag: dbrx
+      models:
+      - model: DBRX Instruct
+        mad_tag: pyt_vllm_dbrx-instruct
+        model_repo: databricks/dbrx-instruct
+        url: https://huggingface.co/databricks/dbrx-instruct
+        precision: float16
+      - model: DBRX Instruct FP8
+        mad_tag: pyt_vllm_dbrx_fp8
+        model_repo: amd/dbrx-instruct-FP8-KV
+        url: https://huggingface.co/amd/dbrx-instruct-FP8-KV
+        precision: float8
+    - group: Google Gemma
+      tag: gemma
+      models:
+      - model: Gemma 2 27B
+        mad_tag: pyt_vllm_gemma-2-27b
+        model_repo: google/gemma-2-27b
+        url: https://huggingface.co/google/gemma-2-27b
+        precision: float16
+    - group: Cohere
+      tag: cohere
+      models:
+      - model: C4AI Command R+ 08-2024
+        mad_tag: pyt_vllm_c4ai-command-r-plus-08-2024
+        model_repo: CohereForAI/c4ai-command-r-plus-08-2024
+        url: https://huggingface.co/CohereForAI/c4ai-command-r-plus-08-2024
+        precision: float16
+      - model: C4AI Command R+ 08-2024 FP8
+        mad_tag: pyt_vllm_command-r-plus_fp8
+        model_repo: amd/c4ai-command-r-plus-FP8-KV
+        url: https://huggingface.co/amd/c4ai-command-r-plus-FP8-KV
+        precision: float8
+    - group: DeepSeek
+      tag: deepseek
+      models:
+      - model: DeepSeek MoE 16B
+        mad_tag: pyt_vllm_deepseek-moe-16b-chat
+        model_repo: deepseek-ai/deepseek-moe-16b-chat
+        url: https://huggingface.co/deepseek-ai/deepseek-moe-16b-chat
+        precision: float16
+    - group: Microsoft Phi
+      tag: phi
+      models:
+      - model: Phi-4
+        mad_tag: pyt_vllm_phi-4
+        model_repo: microsoft/phi-4
+        url: https://huggingface.co/microsoft/phi-4
+    - group: TII Falcon
+      tag: falcon
+      models:
+      - model: Falcon 180B
+        mad_tag: pyt_vllm_falcon-180b
+        model_repo: tiiuae/falcon-180B
+        url: https://huggingface.co/tiiuae/falcon-180B
+        precision: float16

docs/data/how-to/rocm-for-ai/inference/pytorch-inference-benchmark-models.yaml

@@ -31,3 +31,11 @@ pytorch_inference_benchmark:
         model_repo: genmo/mochi-1-preview
         url: https://huggingface.co/genmo/mochi-1-preview
         precision: float16
+    - group: Wan2.1
+      tag: wan
+      models:
+      - model: Wan2.1
+        mad_tag: pyt_wan2.1_inference
+        model_repo: Wan-AI/Wan2.1-T2V-14B
+        url: https://huggingface.co/Wan-AI/Wan2.1-T2V-14B
+        precision: bfloat16

docs/data/how-to/rocm-for-ai/inference/vllm-benchmark-models.yaml

@@ -1,10 +1,11 @@
 vllm_benchmark:
   unified_docker:
     latest:
-      pull_tag: rocm/vllm:rocm6.4.1_vllm_0.9.0.1_20250605
+      # TODO: update me
-      docker_hub_url: https://hub.docker.com/layers/rocm/vllm/rocm6.4.1_vllm_0.9.0.1_20250605/images/sha256-f48beeb3d72663a93c77211eb45273d564451447c097e060befa713d565fa36c
+      pull_tag: rocm/vllm:rocm6.4.1_vllm_0.9.1_20250715
+      docker_hub_url: https://hub.docker.com/layers/rocm/vllm/rocm6.4.1_vllm_0.9.1_20250715/images/sha256-4a429705fa95a58f6d20aceab43b1b76fa769d57f32d5d28bd3f4e030e2a78ea
       rocm_version: 6.4.1
-      vllm_version: 0.9.0.1 (0.9.0.2.dev108+g71faa1880.rocm641)
+      vllm_version: 0.9.1 (0.9.2.dev364+gb432b7a28.rocm641)
       pytorch_version: 2.7.0+gitf717b2a
       hipblaslt_version: 0.15
   model_groups:

docs/data/how-to/rocm-for-ai/training/megatron-lm-benchmark-models.yaml

@@ -1,29 +1,60 @@
-megatron-lm_benchmark:
+dockers:
-  model_groups:
+  - pull_tag: rocm/megatron-lm:v25.6_py312
-    - group: Meta Llama
+    docker_hub_url: https://hub.docker.com/layers/rocm/megatron-lm/v25.6_py312/images/sha256-482ff906532285bceabdf2bda629bd32cb6174d2d07f4243a736378001b28df0
-      tag: llama
+    components:
-      models:
+      ROCm: 6.4.1
+      PyTorch: 2.8.0a0+git7d205b2
+      Python: 3.12
+      Transformer Engine: 2.1.0.dev0+8c4a512
+      hipBLASLt: 393e413
+      Triton: 3.3.0
+      RCCL: 2.23.4.7a84c5d
+    doc_name: Ubuntu 24.04 + Python 3.12
+  - pull_tag: rocm/megatron-lm:v25.6_py310
+    docker_hub_url: https://hub.docker.com/layers/rocm/megatron-lm/v25.6_py310/images/sha256-9627bd9378684fe26cb1a10c7dd817868f553b33402e49b058355b0f095568d6
+    components:
+      ROCm: 6.4.1
+      PyTorch: 2.8.0a0+git7d205b2
+      Python: "3.10"
+      Transformer Engine: 2.1.0.dev0+8c4a512
+      hipBLASLt: 393e413
+      Triton: 3.3.0
+      RCCL: 2.23.4.7a84c5d
+    doc_name: Ubuntu 22.04 + Python 3.10
+model_groups:
+  - group: Meta Llama
+    tag: llama
+    models:
       - model: Llama 3.3 70B
         mad_tag: pyt_megatron_lm_train_llama-3.3-70b
       - model: Llama 3.1 8B
         mad_tag: pyt_megatron_lm_train_llama-3.1-8b
       - model: Llama 3.1 70B
         mad_tag: pyt_megatron_lm_train_llama-3.1-70b
+      - model: Llama 3.1 70B (proxy)
+        mad_tag: pyt_megatron_lm_train_llama-3.1-70b-proxy
       - model: Llama 2 7B
         mad_tag: pyt_megatron_lm_train_llama-2-7b
       - model: Llama 2 70B
         mad_tag: pyt_megatron_lm_train_llama-2-70b
-    - group: DeepSeek
+  - group: DeepSeek
-      tag: deepseek
+    tag: deepseek
-      models:
+    models:
-      - model: DeepSeek-V3
+      - model: DeepSeek-V3 (proxy)
         mad_tag: pyt_megatron_lm_train_deepseek-v3-proxy
       - model: DeepSeek-V2-Lite
         mad_tag: pyt_megatron_lm_train_deepseek-v2-lite-16b
-    - group: Mistral AI
+  - group: Mistral AI
-      tag: mistral
+    tag: mistral
-      models:
+    models:
       - model: Mixtral 8x7B
         mad_tag: pyt_megatron_lm_train_mixtral-8x7b
-      - model: Mixtral 8x22B
+      - model: Mixtral 8x22B (proxy)
         mad_tag: pyt_megatron_lm_train_mixtral-8x22b-proxy
+  - group: Qwen
+    tag: qwen
+    models:
+      - model: Qwen 2.5 7B
+        mad_tag: pyt_megatron_lm_train_qwen2.5-7b
+      - model: Qwen 2.5 72B
+        mad_tag: pyt_megatron_lm_train_qwen2.5-72b

docs/data/how-to/rocm-for-ai/training/previous-versions/megatron-lm-v25.5-benchmark-models.yaml

@@ -0,0 +1,29 @@
+megatron-lm_benchmark:
+  model_groups:
+    - group: Meta Llama
+      tag: llama
+      models:
+      - model: Llama 3.3 70B
+        mad_tag: pyt_megatron_lm_train_llama-3.3-70b
+      - model: Llama 3.1 8B
+        mad_tag: pyt_megatron_lm_train_llama-3.1-8b
+      - model: Llama 3.1 70B
+        mad_tag: pyt_megatron_lm_train_llama-3.1-70b
+      - model: Llama 2 7B
+        mad_tag: pyt_megatron_lm_train_llama-2-7b
+      - model: Llama 2 70B
+        mad_tag: pyt_megatron_lm_train_llama-2-70b
+    - group: DeepSeek
+      tag: deepseek
+      models:
+      - model: DeepSeek-V3
+        mad_tag: pyt_megatron_lm_train_deepseek-v3-proxy
+      - model: DeepSeek-V2-Lite
+        mad_tag: pyt_megatron_lm_train_deepseek-v2-lite-16b
+    - group: Mistral AI
+      tag: mistral
+      models:
+      - model: Mixtral 8x7B
+        mad_tag: pyt_megatron_lm_train_mixtral-8x7b
+      - model: Mixtral 8x22B
+        mad_tag: pyt_megatron_lm_train_mixtral-8x22b-proxy

docs/how-to/gpu-performance/mi300x.rst

@@ -7,21 +7,21 @@ AMD Instinct MI300X performance guides
 **************************************
 
 The following performance guides provide essential guidance on the necessary
-steps to properly :doc:`configure your system for AMD Instinct™ MI300X
+steps to properly `configure your system for AMD Instinct™ MI300X accelerators
-accelerators <../system-optimization/mi300x>`. They include detailed
+<https://instinct.docs.amd.com/projects/amdgpu-docs/en/latest/system-optimization/mi300x.html>`_.
-instructions on system settings and application :doc:`workload tuning
+They include detailed instructions on system settings and application
-<../rocm-for-ai/inference-optimization/workload>` to help you
+:doc:`workload tuning </how-to/rocm-for-ai/inference-optimization/workload>` to
-leverage the maximum capabilities of these accelerators and achieve superior
+help you leverage the maximum capabilities of these accelerators and achieve
-performance.
+superior performance.
 
 * `AMD Instinct MI300X system optimization <https://instinct.docs.amd.com/projects/amdgpu-docs/en/latest/system-optimization/mi300x.html>`__
   covers essential system settings and system management practices to configure
   your AMD Instinct MI300X system for performance.
 
-* :doc:`../rocm-for-ai/inference-optimization/workload` covers steps to
+* :doc:`/how-to/rocm-for-ai/inference-optimization/workload` covers steps to
   optimize the performance of AMD Instinct MI300X series accelerators for HPC
   and deep learning operations.
 
-* :doc:`../rocm-for-ai/inference/vllm-benchmark` introduces a preconfigured
+* :doc:`/how-to/rocm-for-ai/inference/benchmark-docker/vllm` introduces a preconfigured
   environment for LLM inference, designed to help you test performance with
   popular models on AMD Instinct MI300X series accelerators.

docs/how-to/rocm-for-ai/fine-tuning/index.rst

@@ -24,5 +24,3 @@ training, fine-tuning, and inference. It leverages popular machine learning fram
 - :doc:`Fine-tuning and inference <fine-tuning-and-inference>` using a
   :doc:`single-accelerator <single-gpu-fine-tuning-and-inference>` or
   :doc:`multi-accelerator <multi-gpu-fine-tuning-and-inference>` system.
-
-

docs/how-to/rocm-for-ai/index.rst

@@ -6,7 +6,7 @@
 Use ROCm for AI
 **************************
 
-ROCm™ is an open-source software platform that enables high-performance computing and machine learning applications. It features the ability to accelerate training, fine-tuning, and inference for AI application development. With ROCm, you can access the full power of AMD GPUs, which can significantly improve the performance and efficiency of AI workloads.
+ROCm is an open-source software platform that enables high-performance computing and machine learning applications. It features the ability to accelerate training, fine-tuning, and inference for AI application development. With ROCm, you can access the full power of AMD GPUs, which can significantly improve the performance and efficiency of AI workloads.
 
 You can use ROCm to perform distributed training, which enables you to train models across multiple GPUs or nodes simultaneously. Additionally, ROCm supports mixed-precision training, which can help reduce the memory and compute requirements of training workloads. For fine-tuning, ROCm provides access to various algorithms and optimization techniques. In terms of inference, ROCm provides several techniques that can help you optimize your models for deployment, such as quantization, GEMM tuning, and optimization with composable kernel.
  

docs/how-to/rocm-for-ai/inference-optimization/workload.rst

@@ -151,8 +151,8 @@ desired effect. Continuous iteration helps refine the performance gains and
 address any new bottlenecks that may emerge.
 
 ROCm provides a prebuilt optimized Docker image that has everything required to implement
-the tips in this section. It includes ROCm, vLLM, PyTorch, and tuning files in the CSV 
+the LLM inference tips in this section. It includes ROCm, PyTorch, and vLLM.
-format. For more information, see :doc:`../inference/vllm-benchmark`.
+For more information, see :doc:`/how-to/rocm-for-ai/inference/benchmark-docker/vllm`.
 
 .. _mi300x-profiling-tools:
 
@@ -343,9 +343,10 @@ The following performance tips are not *specific* to vLLM -- they are general
 but relevant in this context. You can tune the following vLLM parameters to
 achieve optimal request latency and throughput performance.
 
-* As described in :ref:`mi300x-env-vars`, the environment
+* As described in `Environment variables (MI300X)
-  variable ``HIP_FORCE_DEV_KERNARG`` can improve vLLM performance. Set it to
+  <https://instinct.docs.amd.com/projects/amdgpu-docs/en/latest/system-optimization/mi300x.html#environment-variables>`_,
-  ``export HIP_FORCE_DEV_KERNARG=1``.
+  the environment variable ``HIP_FORCE_DEV_KERNARG`` can improve vLLM
+  performance. Set it to ``export HIP_FORCE_DEV_KERNARG=1``.
 
 * Set the :ref:`RCCL environment variable <mi300x-rccl>` ``NCCL_MIN_NCHANNELS``
   to ``112`` to increase the number of channels on MI300X to potentially improve
@@ -410,9 +411,9 @@ for additional performance tips. :ref:`fine-tuning-llms-vllm` describes vLLM
 usage with ROCm.
 
 ROCm provides a prebuilt optimized Docker image for validating the performance
-of LLM inference with vLLM on the MI300X accelerator. The Docker image includes
+of LLM inference with vLLM on MI300X series accelerators. The Docker image includes
-ROCm, vLLM, PyTorch, and tuning files in the CSV format. For more information,
+ROCm, vLLM, and PyTorch. For more information, see
-see :doc:`../inference/vllm-benchmark`.
+:doc:`/how-to/rocm-for-ai/inference/benchmark-docker/vllm`.
 
 .. _mi300x-vllm-throughput-measurement:
 
@@ -1477,8 +1478,9 @@ following command: ``cat /proc/sys/kernel/numa_balancing`` and
 checking whether the output is ``0``.
 
 If the output is ``1``, you can disable NUMA auto-balancing by running the
-following command: ``sudo sysctl kernel.numa_balancing=0``. For more
+following command: ``sudo sysctl kernel.numa_balancing=0``. For more details,
-details, see :ref:`AMD Instinct MI300X system optimization <mi300x-disable-numa>`.
+see `AMD Instinct MI300X system optimization
+<https://instinct.docs.amd.com/projects/amdgpu-docs/en/latest/system-optimization/mi300x.html#disable-numa-auto-balancing>`_.
 
 .. _mi300x-rccl-disable-acs:
 

docs/how-to/rocm-for-ai/inference/benchmark-docker/previous-versions/vllm-0.4.3.rst

@@ -59,7 +59,7 @@ MI300X accelerator with the prebuilt vLLM Docker image.
 
    To optimize performance, disable automatic NUMA balancing. Otherwise, the GPU
    might hang until the periodic balancing is finalized. For more information,
-   see :ref:`AMD Instinct MI300X system optimization <mi300x-disable-numa>`.
+   see the :ref:`system validation steps <rocm-for-ai-system-optimization>`.
 
    .. code-block:: shell
 
@@ -80,11 +80,11 @@ MI300X accelerator with the prebuilt vLLM Docker image.
 Once setup is complete, you can choose between two options to reproduce the
 benchmark results:
 
--  :ref:`MAD-integrated benchmarking <vllm-benchmark-mad>`
+-  :ref:`MAD-integrated benchmarking <vllm-benchmark-mad-v043>`
 
--  :ref:`Standalone benchmarking <vllm-benchmark-standalone>`
+-  :ref:`Standalone benchmarking <vllm-benchmark-standalone-v043>`
 
-.. _vllm-benchmark-mad:
+.. _vllm-benchmark-mad-v043:
 
 MAD-integrated benchmarking
 ===========================
@@ -112,7 +112,7 @@ model are collected in the following path: ``~/MAD/reports_float16/``
 
 Although the following eight models are pre-configured to collect latency and
 throughput performance data, users can also change the benchmarking parameters.
-Refer to the :ref:`Standalone benchmarking <vllm-benchmark-standalone>` section.
+Refer to the :ref:`Standalone benchmarking <vllm-benchmark-standalone-v043>` section.
 
 Available models
 ----------------
@@ -136,7 +136,7 @@ Available models
 
    * ``pyt_vllm_jais-30b``
 
-.. _vllm-benchmark-standalone:
+.. _vllm-benchmark-standalone-v043:
 
 Standalone benchmarking
 =======================
@@ -167,14 +167,14 @@ Command
 ^^^^^^^^^^^^^^^^^^^^^^^^^
 
 To start the benchmark, use the following command with the appropriate options.
-See :ref:`Options <vllm-benchmark-standalone-options>` for the list of
+See :ref:`Options <vllm-benchmark-standalone-options-v043>` for the list of
 options and their descriptions.
 
 .. code-block:: shell
 
    ./vllm_benchmark_report.sh -s $test_option -m $model_repo -g $num_gpu -d $datatype
 
-See the :ref:`examples <vllm-benchmark-run-benchmark>` for more information.
+See the :ref:`examples <vllm-benchmark-run-benchmark-v043>` for more information.
 
 .. note::
 
@@ -193,7 +193,7 @@ See the :ref:`examples <vllm-benchmark-run-benchmark>` for more information.
       # pass your HF_TOKEN
       export HF_TOKEN=$your_personal_hf_token
 
-.. _vllm-benchmark-standalone-options:
+.. _vllm-benchmark-standalone-options-v043:
 
 Options
 ^^^^^^^^^^^^^^^^^^^^^^^^^
@@ -265,13 +265,13 @@ Options
      - ``float16``
      - Data type
 
-.. _vllm-benchmark-run-benchmark:
+.. _vllm-benchmark-run-benchmark-v043:
 
 Running the benchmark on the MI300X accelerator
 -----------------------------------------------
 
 Here are some examples of running the benchmark with various options.
-See :ref:`Options <vllm-benchmark-standalone-options>` for the list of
+See :ref:`Options <vllm-benchmark-standalone-options-v043>` for the list of
 options and their descriptions.
 
 Latency benchmark example
@@ -322,22 +322,22 @@ Further reading
 ===============
 
 - For application performance optimization strategies for HPC and AI workloads,
-  including inference with vLLM, see :doc:`/how-to/tuning-guides/mi300x/workload`.
+  including inference with vLLM, see :doc:`/how-to/rocm-for-ai/inference-optimization/workload`.
 
 - To learn more about the options for latency and throughput benchmark scripts,
   see `<https://github.com/ROCm/vllm/tree/main/benchmarks>`_.
 
 - To learn more about system settings and management practices to configure your system for
-  MI300X accelerators, see :doc:`/how-to/system-optimization/mi300x`.
+  MI300X series accelerators, see `AMD Instinct MI300X system optimization <https://instinct.docs.amd.com/projects/amdgpu-docs/en/latest/system-optimization/mi300x.html>`_
 
-- To learn how to run LLM models from Hugging Face or your own model, see
+- To learn how to run community models from Hugging Face on AMD GPUs, see
-  :doc:`Using ROCm for AI </how-to/rocm-for-ai/index>`.
+  :doc:`Running models from Hugging Face </how-to/rocm-for-ai/inference/hugging-face-models>`.
 
-- To learn how to optimize inference on LLMs, see
+- To learn how to fine-tune LLMs and optimize inference, see
-  :doc:`Fine-tuning LLMs and inference optimization </how-to/llm-fine-tuning-optimization/index>`.
+  :doc:`Fine-tuning LLMs and inference optimization </how-to/rocm-for-ai/fine-tuning/fine-tuning-and-inference>`.
 
-- For a list of other ready-made Docker images for ROCm, see the
+- For a list of other ready-made Docker images for AI with ROCm, see
-  :doc:`Docker image support matrix <rocm-install-on-linux:reference/docker-image-support-matrix>`.
+  `AMD Infinity Hub <https://www.amd.com/en/developer/resources/infinity-hub.html#f-amd_hub_category=AI%20%26%20ML%20Models>`_.
 
 Previous versions
 =================

docs/how-to/rocm-for-ai/inference/benchmark-docker/previous-versions/vllm-0.6.4.rst

@@ -82,7 +82,7 @@ MI300X accelerator with the prebuilt vLLM Docker image.
 
    To optimize performance, disable automatic NUMA balancing. Otherwise, the GPU
    might hang until the periodic balancing is finalized. For more information,
-   see :ref:`AMD Instinct MI300X system optimization <mi300x-disable-numa>`.
+   see the :ref:`system validation steps <rocm-for-ai-system-optimization>`.
 
    .. code-block:: shell
 
@@ -103,11 +103,11 @@ MI300X accelerator with the prebuilt vLLM Docker image.
 Once setup is complete, you can choose between two options to reproduce the
 benchmark results:
 
--  :ref:`MAD-integrated benchmarking <vllm-benchmark-mad>`
+-  :ref:`MAD-integrated benchmarking <vllm-benchmark-mad-v064>`
 
--  :ref:`Standalone benchmarking <vllm-benchmark-standalone>`
+-  :ref:`Standalone benchmarking <vllm-benchmark-standalone-v064>`
 
-.. _vllm-benchmark-mad:
+.. _vllm-benchmark-mad-v064:
 
 MAD-integrated benchmarking
 ===========================
@@ -135,7 +135,7 @@ model are collected in the following path: ``~/MAD/reports_float16/``.
 
 Although the following models are preconfigured to collect latency and
 throughput performance data, you can also change the benchmarking parameters.
-Refer to the :ref:`Standalone benchmarking <vllm-benchmark-standalone>` section.
+Refer to the :ref:`Standalone benchmarking <vllm-benchmark-standalone-v064>` section.
 
 Available models
 ----------------
@@ -177,7 +177,7 @@ Available models
 
    * ``pyt_vllm_mixtral-8x22b_fp8``
 
-.. _vllm-benchmark-standalone:
+.. _vllm-benchmark-standalone-v064:
 
 Standalone benchmarking
 =======================
@@ -203,14 +203,14 @@ Command
 -------
 
 To start the benchmark, use the following command with the appropriate options.
-See :ref:`Options <vllm-benchmark-standalone-options>` for the list of
+See :ref:`Options <vllm-benchmark-standalone-v064-options>` for the list of
 options and their descriptions.
 
 .. code-block:: shell
 
    ./vllm_benchmark_report.sh -s $test_option -m $model_repo -g $num_gpu -d $datatype
 
-See the :ref:`examples <vllm-benchmark-run-benchmark>` for more information.
+See the :ref:`examples <vllm-benchmark-run-benchmark-v064>` for more information.
 
 .. note::
 
@@ -229,7 +229,7 @@ See the :ref:`examples <vllm-benchmark-run-benchmark>` for more information.
       # pass your HF_TOKEN
       export HF_TOKEN=$your_personal_hf_token
 
-.. _vllm-benchmark-standalone-options:
+.. _vllm-benchmark-standalone-v064-options:
 
 Options
 -------
@@ -330,13 +330,13 @@ Options
      - ``float16`` or ``float8``
      - Data type
 
-.. _vllm-benchmark-run-benchmark:
+.. _vllm-benchmark-run-benchmark-v064:
 
 Running the benchmark on the MI300X accelerator
 -----------------------------------------------
 
 Here are some examples of running the benchmark with various options.
-See :ref:`Options <vllm-benchmark-standalone-options>` for the list of
+See :ref:`Options <vllm-benchmark-standalone-v064-options>` for the list of
 options and their descriptions.
 
 Example 1: latency benchmark
@@ -392,25 +392,22 @@ Further reading
 ===============
 
 - For application performance optimization strategies for HPC and AI workloads,
-  including inference with vLLM, see :doc:`/how-to/tuning-guides/mi300x/workload`.
+  including inference with vLLM, see :doc:`/how-to/rocm-for-ai/inference-optimization/workload`.
 
 - To learn more about the options for latency and throughput benchmark scripts,
   see `<https://github.com/ROCm/vllm/tree/main/benchmarks>`_.
 
 - To learn more about system settings and management practices to configure your system for
-  MI300X accelerators, see :doc:`/how-to/system-optimization/mi300x`.
+  MI300X series accelerators, see `AMD Instinct MI300X system optimization <https://instinct.docs.amd.com/projects/amdgpu-docs/en/latest/system-optimization/mi300x.html>`_
 
-- To learn how to run LLM models from Hugging Face or your own model, see
+- To learn how to run community models from Hugging Face on AMD GPUs, see
-  :doc:`Using ROCm for AI </how-to/rocm-for-ai/index>`.
+  :doc:`Running models from Hugging Face </how-to/rocm-for-ai/inference/hugging-face-models>`.
 
-- To learn how to optimize inference on LLMs, see
+- To learn how to fine-tune LLMs and optimize inference, see
-  :doc:`Fine-tuning LLMs and inference optimization </how-to/llm-fine-tuning-optimization/index>`.
+  :doc:`Fine-tuning LLMs and inference optimization </how-to/rocm-for-ai/fine-tuning/fine-tuning-and-inference>`.
 
-- For a list of other ready-made Docker images for ROCm, see the
+- For a list of other ready-made Docker images for AI with ROCm, see
-  :doc:`Docker image support matrix <rocm-install-on-linux:reference/docker-image-support-matrix>`.
+  `AMD Infinity Hub <https://www.amd.com/en/developer/resources/infinity-hub.html#f-amd_hub_category=AI%20%26%20ML%20Models>`_.
-
-- To compare with the previous version of the ROCm vLLM Docker image for performance validation, refer to
-  `LLM inference performance validation on AMD Instinct MI300X (ROCm 6.2.0) <https://rocm.docs.amd.com/en/docs-6.2.0/how-to/performance-validation/mi300x/vllm-benchmark.html>`_.
 
 Previous versions
 =================

docs/how-to/rocm-for-ai/inference/benchmark-docker/previous-versions/vllm-0.6.6.rst

@@ -31,8 +31,8 @@ accelerator and includes the following components:
 With this Docker image, you can quickly validate the expected inference
 performance numbers for the MI300X accelerator. This topic also provides tips on
 optimizing performance with popular AI models. For more information, see the lists of
-:ref:`available models for MAD-integrated benchmarking <vllm-benchmark-mad-models>`
+:ref:`available models for MAD-integrated benchmarking <vllm-benchmark-mad-v066-models>`
-and :ref:`standalone benchmarking <vllm-benchmark-standalone-options>`.
+and :ref:`standalone benchmarking <vllm-benchmark-standalone-v066-options>`.
 
 .. _vllm-benchmark-vllm:
 
@@ -55,7 +55,7 @@ MI300X accelerator with the prebuilt vLLM Docker image.
 
    To optimize performance, disable automatic NUMA balancing. Otherwise, the GPU
    might hang until the periodic balancing is finalized. For more information,
-   see :ref:`AMD Instinct MI300X system optimization <mi300x-disable-numa>`.
+   see the :ref:`system validation steps <rocm-for-ai-system-optimization>`.
 
    .. code-block:: shell
 
@@ -76,11 +76,11 @@ MI300X accelerator with the prebuilt vLLM Docker image.
 Once the setup is complete, choose between two options to reproduce the
 benchmark results:
 
--  :ref:`MAD-integrated benchmarking <vllm-benchmark-mad>`
+-  :ref:`MAD-integrated benchmarking <vllm-benchmark-mad-v066>`
 
--  :ref:`Standalone benchmarking <vllm-benchmark-standalone>`
+-  :ref:`Standalone benchmarking <vllm-benchmark-standalone-v066>`
 
-.. _vllm-benchmark-mad:
+.. _vllm-benchmark-mad-v066:
 
 MAD-integrated benchmarking
 ===========================
@@ -108,9 +108,9 @@ model are collected in the following path: ``~/MAD/reports_float16/``.
 
 Although the following models are preconfigured to collect latency and
 throughput performance data, you can also change the benchmarking parameters.
-Refer to the :ref:`Standalone benchmarking <vllm-benchmark-standalone>` section.
+Refer to the :ref:`Standalone benchmarking <vllm-benchmark-standalone-v066>` section.
 
-.. _vllm-benchmark-mad-models:
+.. _vllm-benchmark-mad-v066-models:
 
 Available models
 ----------------
@@ -134,10 +134,10 @@ Available models
    * - `Llama 3.2 11B Vision <https://huggingface.co/meta-llama/Llama-3.2-11B-Vision-Instruct>`_
      - ``pyt_vllm_llama-3.2-11b-vision-instruct``
 
-   * - `Llama 2 7B <https://huggingface.co/meta-llama/Llama-2-7b-chat-hf>`_
+   * - `Llama 2 7B <https://huggingface.co/meta-llama/Llama-2-7b-chat-hf>`__
      - ``pyt_vllm_llama-2-7b``
 
-   * - `Llama 2 70B <https://huggingface.co/meta-llama/Llama-2-70b-chat-hf>`_
+   * - `Llama 2 70B <https://huggingface.co/meta-llama/Llama-2-70b-chat-hf>`__
      - ``pyt_vllm_llama-2-70b``
 
    * - `Mixtral MoE 8x7B <https://huggingface.co/mistralai/Mixtral-8x7B-Instruct-v0.1>`_
@@ -194,7 +194,7 @@ Available models
    * - `C4AI Command R+ 08-2024 FP8 <https://huggingface.co/amd/c4ai-command-r-plus-FP8-KV>`_
      - ``pyt_vllm_command-r-plus_fp8``
 
-.. _vllm-benchmark-standalone:
+.. _vllm-benchmark-standalone-v066:
 
 Standalone benchmarking
 =======================
@@ -220,14 +220,14 @@ Command
 -------
 
 To start the benchmark, use the following command with the appropriate options.
-See :ref:`Options <vllm-benchmark-standalone-options>` for the list of
+See :ref:`Options <vllm-benchmark-standalone-v066-options>` for the list of
 options and their descriptions.
 
 .. code-block:: shell
 
    ./vllm_benchmark_report.sh -s $test_option -m $model_repo -g $num_gpu -d $datatype
 
-See the :ref:`examples <vllm-benchmark-run-benchmark>` for more information.
+See the :ref:`examples <vllm-benchmark-run-benchmark-v066>` for more information.
 
 .. note::
 
@@ -246,7 +246,7 @@ See the :ref:`examples <vllm-benchmark-run-benchmark>` for more information.
       # pass your HF_TOKEN
       export HF_TOKEN=$your_personal_hf_token
 
-.. _vllm-benchmark-standalone-options:
+.. _vllm-benchmark-standalone-v066-options:
 
 Options and available models
 ----------------------------
@@ -289,11 +289,11 @@ Options and available models
 
    * -
      - ``meta-llama/Llama-2-7b-chat-hf``
-     - `Llama 2 7B <https://huggingface.co/meta-llama/Llama-2-7b-chat-hf>`_
+     - `Llama 2 7B <https://huggingface.co/meta-llama/Llama-2-7b-chat-hf>`__
 
    * -
      - ``meta-llama/Llama-2-70b-chat-hf``
-     - `Llama 2 7B <https://huggingface.co/meta-llama/Llama-2-70b-chat-hf>`_
+     - `Llama 2 70B <https://huggingface.co/meta-llama/Llama-2-70b-chat-hf>`__
 
    * -
      - ``mistralai/Mixtral-8x7B-Instruct-v0.1``
@@ -375,13 +375,13 @@ Options and available models
      - ``float16`` or ``float8``
      - Data type
 
-.. _vllm-benchmark-run-benchmark:
+.. _vllm-benchmark-run-benchmark-v066:
 
 Running the benchmark on the MI300X accelerator
 -----------------------------------------------
 
 Here are some examples of running the benchmark with various options.
-See :ref:`Options <vllm-benchmark-standalone-options>` for the list of
+See :ref:`Options <vllm-benchmark-standalone-v066-options>` for the list of
 options and their descriptions.
 
 Example 1: latency benchmark
@@ -437,22 +437,22 @@ Further reading
 ===============
 
 - For application performance optimization strategies for HPC and AI workloads,
-  including inference with vLLM, see :doc:`../inference-optimization/workload`.
+  including inference with vLLM, see :doc:`/how-to/rocm-for-ai/inference-optimization/workload`.
 
 - To learn more about the options for latency and throughput benchmark scripts,
   see `<https://github.com/ROCm/vllm/tree/main/benchmarks>`_.
 
 - To learn more about system settings and management practices to configure your system for
-  MI300X accelerators, see :doc:`../../system-optimization/mi300x`.
+  MI300X series accelerators, see `AMD Instinct MI300X system optimization <https://instinct.docs.amd.com/projects/amdgpu-docs/en/latest/system-optimization/mi300x.html>`_
 
-- To learn how to run LLM models from Hugging Face or your own model, see
+- To learn how to run community models from Hugging Face on AMD GPUs, see
-  :doc:`Running models from Hugging Face <hugging-face-models>`.
+  :doc:`Running models from Hugging Face </how-to/rocm-for-ai/inference/hugging-face-models>`.
 
-- To learn how to optimize inference on LLMs, see
+- To learn how to fine-tune LLMs and optimize inference, see
-  :doc:`Inference optimization <../inference-optimization/index>`.
+  :doc:`Fine-tuning LLMs and inference optimization </how-to/rocm-for-ai/fine-tuning/fine-tuning-and-inference>`.
 
-- To learn how to fine-tune LLMs, see
+- For a list of other ready-made Docker images for AI with ROCm, see
-  :doc:`Fine-tuning LLMs <../fine-tuning/index>`.
+  `AMD Infinity Hub <https://www.amd.com/en/developer/resources/infinity-hub.html#f-amd_hub_category=AI%20%26%20ML%20Models>`_.
 
 Previous versions
 =================

docs/how-to/rocm-for-ai/inference/benchmark-docker/previous-versions/vllm-0.7.3-20250325.rst

@@ -36,10 +36,10 @@ vLLM inference performance testing
    * `hipBLASLt {{ unified_docker.hipblaslt_version }} <https://github.com/ROCm/hipBLASLt>`_
 
    With this Docker image, you can quickly test the :ref:`expected
-   inference performance numbers <vllm-benchmark-performance-measurements>` for
+   inference performance numbers <vllm-benchmark-performance-measurements-v073>` for
    MI300X series accelerators.
 
-   .. _vllm-benchmark-available-models:
+   .. _vllm-benchmark-available-models-v073:
 
    Available models
    ================
@@ -95,7 +95,7 @@ vLLM inference performance testing
       See :ref:`fine-tuning-llms-vllm` and :ref:`mi300x-vllm-optimization` for
       more information.
 
-   .. _vllm-benchmark-performance-measurements:
+   .. _vllm-benchmark-performance-measurements-v073:
 
    Performance measurements
    ========================
@@ -109,7 +109,7 @@ vLLM inference performance testing
 
       The performance data presented in
       `Performance results with AMD ROCm software <https://www.amd.com/en/developer/resources/rocm-hub/dev-ai/performance-results.html>`_
-      only reflects the :doc:`latest version of this inference benchmarking environment <../vllm>`_.
+      only reflects the :doc:`latest version of this inference benchmarking environment <../vllm>`.
       The listed measurements should not be interpreted as the peak performance achievable by AMD Instinct MI325X and MI300X accelerators or ROCm software.
 
    Advanced features and known issues
@@ -130,7 +130,7 @@ vLLM inference performance testing
 
       To optimize performance, disable automatic NUMA balancing. Otherwise, the GPU
       might hang until the periodic balancing is finalized. For more information,
-      see :ref:`AMD Instinct MI300X system optimization <mi300x-disable-numa>`.
+      see the :ref:`system validation steps <rocm-for-ai-system-optimization>`.
 
       .. code-block:: shell
 
@@ -154,7 +154,7 @@ vLLM inference performance testing
    Once the setup is complete, choose between two options to reproduce the
    benchmark results:
 
-   .. _vllm-benchmark-mad:
+   .. _vllm-benchmark-mad-v073:
 
    {% for model_group in model_groups %}
       {% for model in model_group.models %}
@@ -175,7 +175,7 @@ vLLM inference performance testing
                pip install -r requirements.txt
 
             Use this command to run the performance benchmark test on the `{{model.model}} <{{ model.url }}>`_ model
-            using one GPU with the ``{{model.precision}}`` data type on the host machine.
+            using one GPU with the :literal:`{{model.precision}}` data type on the host machine.
 
             .. code-block:: shell
 
@@ -186,7 +186,7 @@ vLLM inference performance testing
             ``container_ci-{{model.mad_tag}}``. The latency and throughput reports of the
             model are collected in the following path: ``~/MAD/reports_{{model.precision}}/``.
 
-            Although the :ref:`available models <vllm-benchmark-available-models>` are preconfigured
+            Although the :ref:`available models <vllm-benchmark-available-models-v073>` are preconfigured
             to collect latency and throughput performance data, you can also change the benchmarking
             parameters. See the standalone benchmarking tab for more information.
 
@@ -264,7 +264,7 @@ vLLM inference performance testing
 
             * Latency benchmark
 
-              Use this command to benchmark the latency of the {{model.model}} model on eight GPUs with the ``{{model.precision}}`` data type.
+              Use this command to benchmark the latency of the {{model.model}} model on eight GPUs with the :literal:`{{model.precision}}` data type.
 
               .. code-block::
 
@@ -274,7 +274,7 @@ vLLM inference performance testing
 
             * Throughput benchmark
 
-              Use this command to throughput the latency of the {{model.model}} model on eight GPUs with the ``{{model.precision}}`` data type.
+              Use this command to throughput the latency of the {{model.model}} model on eight GPUs with the :literal:`{{model.precision}}` data type.
 
               .. code-block:: shell
 
@@ -305,22 +305,22 @@ Further reading
 ===============
 
 - For application performance optimization strategies for HPC and AI workloads,
-  including inference with vLLM, see :doc:`../inference-optimization/workload`.
+  including inference with vLLM, see :doc:`/how-to/rocm-for-ai/inference-optimization/workload`.
 
 - To learn more about the options for latency and throughput benchmark scripts,
   see `<https://github.com/ROCm/vllm/tree/main/benchmarks>`_.
 
 - To learn more about system settings and management practices to configure your system for
-  MI300X accelerators, see `AMD Instinct MI300X system optimization <https://instinct.docs.amd.com/projects/amdgpu-docs/en/latest/system-optimization/mi300x.html>`_
+  MI300X series accelerators, see `AMD Instinct MI300X system optimization <https://instinct.docs.amd.com/projects/amdgpu-docs/en/latest/system-optimization/mi300x.html>`_
 
-- To learn how to run LLM models from Hugging Face or your own model, see
+- To learn how to run community models from Hugging Face on AMD GPUs, see
-  :doc:`Running models from Hugging Face <hugging-face-models>`.
+  :doc:`Running models from Hugging Face </how-to/rocm-for-ai/inference/hugging-face-models>`.
 
-- To learn how to optimize inference on LLMs, see
+- To learn how to fine-tune LLMs and optimize inference, see
-  :doc:`Inference optimization <../inference-optimization/index>`.
+  :doc:`Fine-tuning LLMs and inference optimization </how-to/rocm-for-ai/fine-tuning/fine-tuning-and-inference>`.
 
-- To learn how to fine-tune LLMs, see
+- For a list of other ready-made Docker images for AI with ROCm, see
-  :doc:`Fine-tuning LLMs <../fine-tuning/index>`.
+  `AMD Infinity Hub <https://www.amd.com/en/developer/resources/infinity-hub.html#f-amd_hub_category=AI%20%26%20ML%20Models>`_.
 
 Previous versions
 =================

docs/how-to/rocm-for-ai/inference/benchmark-docker/previous-versions/vllm-0.8.3-20250415.rst

@@ -1,3 +1,5 @@
+:orphan:
+
 .. meta::
    :description: Learn how to validate LLM inference performance on MI300X accelerators using AMD MAD and the
                  ROCm vLLM Docker image.
@@ -29,10 +31,10 @@ vLLM inference performance testing
    * `hipBLASLt {{ unified_docker.hipblaslt_version }} <https://github.com/ROCm/hipBLASLt>`_
 
    With this Docker image, you can quickly test the :ref:`expected
-   inference performance numbers <vllm-benchmark-performance-measurements>` for
+   inference performance numbers <vllm-benchmark-performance-measurements-v083>` for
    MI300X series accelerators.
 
-   .. _vllm-benchmark-available-models:
+   .. _vllm-benchmark-available-models-v083:
 
    Supported models
    ================
@@ -88,7 +90,7 @@ vLLM inference performance testing
       See :ref:`fine-tuning-llms-vllm` and :ref:`mi300x-vllm-optimization` for
       more information.
 
-   .. _vllm-benchmark-performance-measurements:
+   .. _vllm-benchmark-performance-measurements-v083:
 
    Performance measurements
    ========================
@@ -102,7 +104,7 @@ vLLM inference performance testing
 
       The performance data presented in
       `Performance results with AMD ROCm software <https://www.amd.com/en/developer/resources/rocm-hub/dev-ai/performance-results.html>`_
-      only reflects the :doc:`latest version of this inference benchmarking environment <../vllm>`_.
+      only reflects the :doc:`latest version of this inference benchmarking environment <../vllm>`.
       The listed measurements should not be interpreted as the peak performance achievable by AMD Instinct MI325X and MI300X accelerators or ROCm software.
 
    Advanced features and known issues
@@ -170,7 +172,7 @@ vLLM inference performance testing
                pip install -r requirements.txt
 
             Use this command to run the performance benchmark test on the `{{model.model}} <{{ model.url }}>`_ model
-            using one GPU with the ``{{model.precision}}`` data type on the host machine.
+            using one GPU with the :literal:`{{model.precision}}` data type on the host machine.
 
             .. code-block:: shell
 
@@ -181,7 +183,7 @@ vLLM inference performance testing
             ``container_ci-{{model.mad_tag}}``. The latency and throughput reports of the
             model are collected in the following path: ``~/MAD/reports_{{model.precision}}/``.
 
-            Although the :ref:`available models <vllm-benchmark-available-models>` are preconfigured
+            Although the :ref:`available models <vllm-benchmark-available-models-v083>` are preconfigured
             to collect latency and throughput performance data, you can also change the benchmarking
             parameters. See the standalone benchmarking tab for more information.
 
@@ -278,7 +280,7 @@ vLLM inference performance testing
 
             * Latency benchmark
 
-              Use this command to benchmark the latency of the {{model.model}} model on eight GPUs with ``{{model.precision}}`` precision.
+              Use this command to benchmark the latency of the {{model.model}} model on eight GPUs with :literal:`{{model.precision}}` precision.
 
               .. code-block::
 
@@ -288,7 +290,7 @@ vLLM inference performance testing
 
             * Throughput benchmark
 
-              Use this command to benchmark the throughput of the {{model.model}} model on eight GPUs with ``{{model.precision}}`` precision.
+              Use this command to benchmark the throughput of the {{model.model}} model on eight GPUs with :literal:`{{model.precision}}` precision.
 
               .. code-block:: shell
 
@@ -319,22 +321,22 @@ Further reading
 ===============
 
 - For application performance optimization strategies for HPC and AI workloads,
-  including inference with vLLM, see :doc:`../inference-optimization/workload`.
+  including inference with vLLM, see :doc:`/how-to/rocm-for-ai/inference-optimization/workload`.
 
 - To learn more about the options for latency and throughput benchmark scripts,
   see `<https://github.com/ROCm/vllm/tree/main/benchmarks>`_.
 
 - To learn more about system settings and management practices to configure your system for
-  MI300X accelerators, see `AMD Instinct MI300X system optimization <https://instinct.docs.amd.com/projects/amdgpu-docs/en/latest/system-optimization/mi300x.html>`_
+  MI300X series accelerators, see `AMD Instinct MI300X system optimization <https://instinct.docs.amd.com/projects/amdgpu-docs/en/latest/system-optimization/mi300x.html>`_
 
-- To learn how to run LLM models from Hugging Face or your own model, see
+- To learn how to run community models from Hugging Face on AMD GPUs, see
-  :doc:`Running models from Hugging Face <hugging-face-models>`.
+  :doc:`Running models from Hugging Face </how-to/rocm-for-ai/inference/hugging-face-models>`.
 
-- To learn how to optimize inference on LLMs, see
+- To learn how to fine-tune LLMs and optimize inference, see
-  :doc:`Inference optimization <../inference-optimization/index>`.
+  :doc:`Fine-tuning LLMs and inference optimization </how-to/rocm-for-ai/fine-tuning/fine-tuning-and-inference>`.
 
-- To learn how to fine-tune LLMs, see
+- For a list of other ready-made Docker images for AI with ROCm, see
-  :doc:`Fine-tuning LLMs <../fine-tuning/index>`.
+  `AMD Infinity Hub <https://www.amd.com/en/developer/resources/infinity-hub.html#f-amd_hub_category=AI%20%26%20ML%20Models>`_.
 
 Previous versions
 =================

docs/how-to/rocm-for-ai/inference/benchmark-docker/previous-versions/vllm-0.8.5-20250513.rst

@@ -36,10 +36,10 @@ vLLM inference performance testing
    * `hipBLASLt {{ unified_docker.hipblaslt_version }} <https://github.com/ROCm/hipBLASLt>`_
 
    With this Docker image, you can quickly test the :ref:`expected
-   inference performance numbers <vllm-benchmark-performance-measurements>` for
+   inference performance numbers <vllm-benchmark-performance-measurements-v085-20250513>` for
    MI300X series accelerators.
 
-   .. _vllm-benchmark-available-models:
+   .. _vllm-benchmark-available-models-v085-20250513:
 
    Supported models
    ================
@@ -99,7 +99,7 @@ vLLM inference performance testing
       See :ref:`fine-tuning-llms-vllm` and :ref:`mi300x-vllm-optimization` for
       more information.
 
-   .. _vllm-benchmark-performance-measurements:
+   .. _vllm-benchmark-performance-measurements-v085-20250513:
 
    Performance measurements
    ========================
@@ -113,7 +113,7 @@ vLLM inference performance testing
 
       The performance data presented in
       `Performance results with AMD ROCm software <https://www.amd.com/en/developer/resources/rocm-hub/dev-ai/performance-results.html>`_
-      only reflects the :doc:`latest version of this inference benchmarking environment <../vllm>`_.
+      only reflects the :doc:`latest version of this inference benchmarking environment <../vllm>`.
       The listed measurements should not be interpreted as the peak performance achievable by AMD Instinct MI325X and MI300X accelerators or ROCm software.
 
    Advanced features and known issues
@@ -181,7 +181,7 @@ vLLM inference performance testing
                pip install -r requirements.txt
 
             Use this command to run the performance benchmark test on the `{{model.model}} <{{ model.url }}>`_ model
-            using one GPU with the ``{{model.precision}}`` data type on the host machine.
+            using one GPU with the :literal:`{{model.precision}}` data type on the host machine.
 
             .. code-block:: shell
 
@@ -192,7 +192,7 @@ vLLM inference performance testing
             ``container_ci-{{model.mad_tag}}``. The latency and throughput reports of the
             model are collected in the following path: ``~/MAD/reports_{{model.precision}}/``.
 
-            Although the :ref:`available models <vllm-benchmark-available-models>` are preconfigured
+            Although the :ref:`available models <vllm-benchmark-available-models-v085-20250513>` are preconfigured
             to collect latency and throughput performance data, you can also change the benchmarking
             parameters. See the standalone benchmarking tab for more information.
 
@@ -289,7 +289,7 @@ vLLM inference performance testing
 
             * Latency benchmark
 
-              Use this command to benchmark the latency of the {{model.model}} model on eight GPUs with ``{{model.precision}}`` precision.
+              Use this command to benchmark the latency of the {{model.model}} model on eight GPUs with :literal:`{{model.precision}}` precision.
 
               .. code-block::
 
@@ -299,7 +299,7 @@ vLLM inference performance testing
 
             * Throughput benchmark
 
-              Use this command to benchmark the throughput of the {{model.model}} model on eight GPUs with ``{{model.precision}}`` precision.
+              Use this command to benchmark the throughput of the {{model.model}} model on eight GPUs with :literal:`{{model.precision}}` precision.
 
               .. code-block:: shell
 
@@ -333,19 +333,19 @@ Further reading
   see `<https://github.com/ROCm/vllm/tree/main/benchmarks>`_.
 
 - To learn more about system settings and management practices to configure your system for
-  MI300X accelerators, see `AMD Instinct MI300X system optimization <https://instinct.docs.amd.com/projects/amdgpu-docs/en/latest/system-optimization/mi300x.html>`_
+  MI300X series accelerators, see `AMD Instinct MI300X system optimization <https://instinct.docs.amd.com/projects/amdgpu-docs/en/latest/system-optimization/mi300x.html>`_
 
 - For application performance optimization strategies for HPC and AI workloads,
-  including inference with vLLM, see :doc:`../../../inference-optimization/workload`.
+  including inference with vLLM, see :doc:`/how-to/rocm-for-ai/inference-optimization/workload`.
 
-- To learn how to run LLM models from Hugging Face or your own model, see
+- To learn how to run community models from Hugging Face on AMD GPUs, see
-  :doc:`Running models from Hugging Face <../../hugging-face-models>`.
+  :doc:`Running models from Hugging Face </how-to/rocm-for-ai/inference/hugging-face-models>`.
 
-- To learn how to optimize inference on LLMs, see
+- To learn how to fine-tune LLMs and optimize inference, see
-  :doc:`Inference optimization <../../../inference-optimization/index>`.
+  :doc:`Fine-tuning LLMs and inference optimization </how-to/rocm-for-ai/fine-tuning/fine-tuning-and-inference>`.
 
-- To learn how to fine-tune LLMs, see
+- For a list of other ready-made Docker images for AI with ROCm, see
-  :doc:`Fine-tuning LLMs <../../../fine-tuning/index>`.
+  `AMD Infinity Hub <https://www.amd.com/en/developer/resources/infinity-hub.html#f-amd_hub_category=AI%20%26%20ML%20Models>`_.
 
 Previous versions
 =================

docs/how-to/rocm-for-ai/inference/benchmark-docker/previous-versions/vllm-0.8.5-20250521.rst

@@ -36,10 +36,10 @@ vLLM inference performance testing
    * `hipBLASLt {{ unified_docker.hipblaslt_version }} <https://github.com/ROCm/hipBLASLt>`_
 
    With this Docker image, you can quickly test the :ref:`expected
-   inference performance numbers <vllm-benchmark-performance-measurements>` for
+   inference performance numbers <vllm-benchmark-performance-measurements-v085-20250521>` for
    MI300X series accelerators.
 
-   .. _vllm-benchmark-available-models:
+   .. _vllm-benchmark-available-models-v085-20250521:
 
    Supported models
    ================
@@ -99,7 +99,7 @@ vLLM inference performance testing
       See :ref:`fine-tuning-llms-vllm` and :ref:`mi300x-vllm-optimization` for
       more information.
 
-   .. _vllm-benchmark-performance-measurements:
+   .. _vllm-benchmark-performance-measurements-v085-20250521:
 
    Performance measurements
    ========================
@@ -181,7 +181,7 @@ vLLM inference performance testing
                pip install -r requirements.txt
 
             Use this command to run the performance benchmark test on the `{{model.model}} <{{ model.url }}>`_ model
-            using one GPU with the ``{{model.precision}}`` data type on the host machine.
+            using one GPU with the :literal:`{{model.precision}}` data type on the host machine.
 
             .. code-block:: shell
 
@@ -192,7 +192,7 @@ vLLM inference performance testing
             ``container_ci-{{model.mad_tag}}``. The latency and throughput reports of the
             model are collected in the following path: ``~/MAD/reports_{{model.precision}}/``.
 
-            Although the :ref:`available models <vllm-benchmark-available-models>` are preconfigured
+            Although the :ref:`available models <vllm-benchmark-available-models-v085-20250521>` are preconfigured
             to collect latency and throughput performance data, you can also change the benchmarking
             parameters. See the standalone benchmarking tab for more information.
 
@@ -289,7 +289,7 @@ vLLM inference performance testing
 
             * Latency benchmark
 
-              Use this command to benchmark the latency of the {{model.model}} model on eight GPUs with ``{{model.precision}}`` precision.
+              Use this command to benchmark the latency of the {{model.model}} model on eight GPUs with :literal:`{{model.precision}}` precision.
 
               .. code-block::
 
@@ -299,7 +299,7 @@ vLLM inference performance testing
 
             * Throughput benchmark
 
-              Use this command to benchmark the throughput of the {{model.model}} model on eight GPUs with ``{{model.precision}}`` precision.
+              Use this command to benchmark the throughput of the {{model.model}} model on eight GPUs with :literal:`{{model.precision}}` precision.
 
               .. code-block:: shell
 
@@ -333,22 +333,23 @@ Further reading
   see `<https://github.com/ROCm/vllm/tree/main/benchmarks>`_.
 
 - To learn more about system settings and management practices to configure your system for
-  MI300X accelerators, see `AMD Instinct MI300X system optimization <https://instinct.docs.amd.com/projects/amdgpu-docs/en/latest/system-optimization/mi300x.html>`_
+  MI300X series accelerators, see `AMD Instinct MI300X system optimization <https://instinct.docs.amd.com/projects/amdgpu-docs/en/latest/system-optimization/mi300x.html>`_
 
 - For application performance optimization strategies for HPC and AI workloads,
-  including inference with vLLM, see :doc:`../../inference-optimization/workload`.
+  including inference with vLLM, see :doc:`/how-to/rocm-for-ai/inference-optimization/workload`.
 
-- To learn how to run LLM models from Hugging Face or your own model, see
+- To learn how to run community models from Hugging Face on AMD GPUs, see
-  :doc:`Running models from Hugging Face <../hugging-face-models>`.
+  :doc:`Running models from Hugging Face </how-to/rocm-for-ai/inference/hugging-face-models>`.
 
-- To learn how to optimize inference on LLMs, see
+- To learn how to fine-tune LLMs and optimize inference, see
-  :doc:`Inference optimization <../../inference-optimization/index>`.
+  :doc:`Fine-tuning LLMs and inference optimization </how-to/rocm-for-ai/fine-tuning/fine-tuning-and-inference>`.
 
-- To learn how to fine-tune LLMs, see
+- For a list of other ready-made Docker images for AI with ROCm, see
-  :doc:`Fine-tuning LLMs <../../fine-tuning/index>`.
+  `AMD Infinity Hub <https://www.amd.com/en/developer/resources/infinity-hub.html#f-amd_hub_category=AI%20%26%20ML%20Models>`_.
 
 Previous versions
 =================
 
 See :doc:`vllm-history` to find documentation for previous releases
 of the ``ROCm/vllm`` Docker image.
+

docs/how-to/rocm-for-ai/inference/benchmark-docker/previous-versions/vllm-0.9.0.1-20250605.rst

@@ -0,0 +1,353 @@
+:orphan:
+
+.. meta::
+   :description: Learn how to validate LLM inference performance on MI300X accelerators using AMD MAD and the
+                 ROCm vLLM Docker image.
+   :keywords: model, MAD, automation, dashboarding, validate
+
+**********************************
+vLLM inference performance testing
+**********************************
+
+.. caution::
+
+   This documentation does not reflect the latest version of ROCm vLLM
+   inference performance documentation. See :doc:`../vllm` for the latest version.
+
+.. _vllm-benchmark-unified-docker:
+
+.. datatemplate:yaml:: /data/how-to/rocm-for-ai/inference/previous-versions/vllm_0.9.0.1_20250605-benchmark-models.yaml
+
+   {% set unified_docker = data.vllm_benchmark.unified_docker.latest %}
+   {% set model_groups = data.vllm_benchmark.model_groups %}
+
+   The `ROCm vLLM Docker <{{ unified_docker.docker_hub_url }}>`_ image offers
+   a prebuilt, optimized environment for validating large language model (LLM)
+   inference performance on AMD Instinct™ MI300X series accelerators. This ROCm vLLM
+   Docker image integrates vLLM and PyTorch tailored specifically for MI300X series
+   accelerators and includes the following components:
+
+   * `ROCm {{ unified_docker.rocm_version }} <https://github.com/ROCm/ROCm>`_
+
+   * `vLLM {{ unified_docker.vllm_version }} <https://docs.vllm.ai/en/latest>`_
+
+   * `PyTorch {{ unified_docker.pytorch_version }} <https://github.com/ROCm/pytorch.git>`_
+
+   * `hipBLASLt {{ unified_docker.hipblaslt_version }} <https://github.com/ROCm/hipBLASLt>`_
+
+   With this Docker image, you can quickly test the :ref:`expected
+   inference performance numbers <vllm-benchmark-performance-measurements-v0901-20250605>` for
+   MI300X series accelerators.
+
+   .. _vllm-benchmark-available-models-v0901-20250605:
+
+   Supported models
+   ================
+
+   The following models are supported for inference performance benchmarking
+   with vLLM and ROCm. Some instructions, commands, and recommendations in this
+   documentation might vary by model -- select one to get started.
+
+   .. raw:: html
+
+      <div id="vllm-benchmark-ud-params-picker" class="container-fluid">
+        <div class="row">
+          <div class="col-2 me-2 model-param-head">Model group</div>
+          <div class="row col-10">
+   {% for model_group in model_groups %}
+            <div class="col-3 model-param" data-param-k="model-group" data-param-v="{{ model_group.tag }}" tabindex="0">{{ model_group.group }}</div>
+   {% endfor %}
+          </div>
+        </div>
+
+        <div class="row mt-1">
+          <div class="col-2 me-2 model-param-head">Model</div>
+          <div class="row col-10">
+   {% for model_group in model_groups %}
+      {% set models = model_group.models %}
+      {% for model in models %}
+         {% if models|length % 3 == 0 %}
+            <div class="col-4 model-param" data-param-k="model" data-param-v="{{ model.mad_tag }}" data-param-group="{{ model_group.tag }}" tabindex="0">{{ model.model }}</div>
+         {% else %}
+            <div class="col-6 model-param" data-param-k="model" data-param-v="{{ model.mad_tag }}" data-param-group="{{ model_group.tag }}" tabindex="0">{{ model.model }}</div>
+         {% endif %}
+      {% endfor %}
+   {% endfor %}
+          </div>
+        </div>
+      </div>
+
+   .. _vllm-benchmark-vllm:
+
+   {% for model_group in model_groups %}
+      {% for model in model_group.models %}
+
+   .. container:: model-doc {{model.mad_tag}}
+
+      .. note::
+
+         See the `{{ model.model }} model card on Hugging Face <{{ model.url }}>`_ to learn more about your selected model.
+         Some models require access authorization prior to use via an external license agreement through a third party.
+
+      {% endfor %}
+   {% endfor %}
+
+   .. note::
+
+      vLLM is a toolkit and library for LLM inference and serving. AMD implements
+      high-performance custom kernels and modules in vLLM to enhance performance.
+      See :ref:`fine-tuning-llms-vllm` and :ref:`mi300x-vllm-optimization` for
+      more information.
+
+   .. _vllm-benchmark-performance-measurements-v0901-20250605:
+
+   Performance measurements
+   ========================
+
+   To evaluate performance, the
+   `Performance results with AMD ROCm software <https://www.amd.com/en/developer/resources/rocm-hub/dev-ai/performance-results.html>`_
+   page provides reference throughput and latency measurements for inferencing popular AI models.
+
+   .. important::
+
+      The performance data presented in
+      `Performance results with AMD ROCm software <https://www.amd.com/en/developer/resources/rocm-hub/dev-ai/performance-results.html>`_
+      only reflects the latest version of this inference benchmarking environment.
+      The listed measurements should not be interpreted as the peak performance achievable by AMD Instinct MI325X and MI300X accelerators or ROCm software.
+
+   Advanced features and known issues
+   ==================================
+
+   For information on experimental features and known issues related to ROCm optimization efforts on vLLM,
+   see the developer's guide at `<https://github.com/ROCm/vllm/tree/7bb0618b1fe725b7d4fad9e525aa44da12c94a8b/docs/dev-docker>`__.
+
+   System validation
+   =================
+
+   Before running AI workloads, it's important to validate that your AMD hardware is configured
+   correctly and performing optimally.
+
+   To optimize performance, disable automatic NUMA balancing. Otherwise, the GPU
+   might hang until the periodic balancing is finalized. For more information,
+   see the :ref:`system validation steps <rocm-for-ai-system-optimization>`.
+
+   .. code-block:: shell
+
+      # disable automatic NUMA balancing
+      sh -c 'echo 0 > /proc/sys/kernel/numa_balancing'
+      # check if NUMA balancing is disabled (returns 0 if disabled)
+      cat /proc/sys/kernel/numa_balancing
+      0
+
+   To test for optimal performance, consult the recommended :ref:`System health benchmarks
+   <rocm-for-ai-system-health-bench>`. This suite of tests will help you verify and fine-tune your
+   system's configuration.
+
+   Pull the Docker image
+   =====================
+
+   Download the `ROCm vLLM Docker image <{{ unified_docker.docker_hub_url }}>`_.
+   Use the following command to pull the Docker image from Docker Hub.
+
+   .. code-block:: shell
+
+      docker pull {{ unified_docker.pull_tag }}
+
+   Benchmarking
+   ============
+
+   Once the setup is complete, choose between two options to reproduce the
+   benchmark results:
+
+   .. _vllm-benchmark-mad:
+
+   {% for model_group in model_groups %}
+      {% for model in model_group.models %}
+
+   .. container:: model-doc {{model.mad_tag}}
+
+      .. tab-set::
+
+         .. tab-item:: MAD-integrated benchmarking
+
+            Clone the ROCm Model Automation and Dashboarding (`<https://github.com/ROCm/MAD>`__) repository to a local
+            directory and install the required packages on the host machine.
+
+            .. code-block:: shell
+
+               git clone https://github.com/ROCm/MAD
+               cd MAD
+               pip install -r requirements.txt
+
+            Use this command to run the performance benchmark test on the `{{model.model}} <{{ model.url }}>`_ model
+            using one GPU with the :literal:`{{model.precision}}` data type on the host machine.
+
+            .. code-block:: shell
+
+               export MAD_SECRETS_HFTOKEN="your personal Hugging Face token to access gated models"
+               python3 tools/run_models.py --tags {{model.mad_tag}} --keep-model-dir --live-output --timeout 28800
+
+            MAD launches a Docker container with the name
+            ``container_ci-{{model.mad_tag}}``. The latency and throughput reports of the
+            model are collected in the following path: ``~/MAD/reports_{{model.precision}}/``.
+
+            Although the :ref:`available models <vllm-benchmark-available-models-v0901-20250605>` are preconfigured
+            to collect latency and throughput performance data, you can also change the benchmarking
+            parameters. See the standalone benchmarking tab for more information.
+
+            {% if model.tunableop %}
+
+            .. note::
+
+               For improved performance, consider enabling :ref:`PyTorch TunableOp <mi300x-tunableop>`.
+               TunableOp automatically explores different implementations and configurations of certain PyTorch
+               operators to find the fastest one for your hardware.
+
+               By default, ``{{model.mad_tag}}`` runs with TunableOp disabled
+               (see
+               `<https://github.com/ROCm/MAD/blob/develop/models.json>`__). To
+               enable it, edit the default run behavior in the ``models.json``
+               configuration before running inference -- update the model's run
+               ``args`` by changing ``--tunableop off`` to ``--tunableop on``.
+
+               Enabling TunableOp triggers a two-pass run -- a warm-up followed by the performance-collection run.
+
+            {% endif %}
+
+         .. tab-item:: Standalone benchmarking
+
+            Run the vLLM benchmark tool independently by starting the
+            `Docker container <{{ unified_docker.docker_hub_url }}>`_
+            as shown in the following snippet.
+
+            .. code-block::
+
+               docker pull {{ unified_docker.pull_tag }}
+               docker run -it --device=/dev/kfd --device=/dev/dri --group-add video --shm-size 16G --security-opt seccomp=unconfined --security-opt apparmor=unconfined --cap-add=SYS_PTRACE -v $(pwd):/workspace --env HUGGINGFACE_HUB_CACHE=/workspace --name test {{ unified_docker.pull_tag }}
+
+            In the Docker container, clone the ROCm MAD repository and navigate to the
+            benchmark scripts directory at ``~/MAD/scripts/vllm``.
+
+            .. code-block::
+
+               git clone https://github.com/ROCm/MAD
+               cd MAD/scripts/vllm
+
+            To start the benchmark, use the following command with the appropriate options.
+
+            .. code-block::
+
+               ./vllm_benchmark_report.sh -s $test_option -m {{model.model_repo}} -g $num_gpu -d {{model.precision}}
+
+            .. list-table::
+               :header-rows: 1
+               :align: center
+
+               * - Name
+                 - Options
+                 - Description
+
+               * - ``$test_option``
+                 - latency
+                 - Measure decoding token latency
+
+               * -
+                 - throughput
+                 - Measure token generation throughput
+
+               * -
+                 - all
+                 - Measure both throughput and latency
+
+               * - ``$num_gpu``
+                 - 1 or 8
+                 - Number of GPUs
+
+               * - ``$datatype``
+                 - ``float16`` or ``float8``
+                 - Data type
+
+            .. note::
+
+               The input sequence length, output sequence length, and tensor parallel (TP) are
+               already configured. You don't need to specify them with this script.
+
+            .. note::
+
+               If you encounter the following error, pass your access-authorized Hugging
+               Face token to the gated models.
+
+               .. code-block::
+
+                  OSError: You are trying to access a gated repo.
+
+                  # pass your HF_TOKEN
+                  export HF_TOKEN=$your_personal_hf_token
+
+            Here are some examples of running the benchmark with various options.
+
+            * Latency benchmark
+
+              Use this command to benchmark the latency of the {{model.model}} model on eight GPUs with :literal:`{{model.precision}}` precision.
+
+              .. code-block::
+
+                 ./vllm_benchmark_report.sh -s latency -m {{model.model_repo}} -g 8 -d {{model.precision}}
+
+              Find the latency report at ``./reports_{{model.precision}}_vllm_rocm{{unified_docker.rocm_version}}/summary/{{model.model_repo.split('/', 1)[1] if '/' in model.model_repo else model.model_repo}}_latency_report.csv``.
+
+            * Throughput benchmark
+
+              Use this command to benchmark the throughput of the {{model.model}} model on eight GPUs with :literal:`{{model.precision}}` precision.
+
+              .. code-block:: shell
+
+                 ./vllm_benchmark_report.sh -s throughput -m {{model.model_repo}} -g 8 -d {{model.precision}}
+
+              Find the throughput report at ``./reports_{{model.precision}}_vllm_rocm{{unified_docker.rocm_version}}/summary/{{model.model_repo.split('/', 1)[1] if '/' in model.model_repo else model.model_repo}}_throughput_report.csv``.
+
+            .. raw:: html
+
+               <style>
+               mjx-container[jax="CHTML"][display="true"] {
+                  text-align: left;
+                  margin: 0;
+               }
+               </style>
+
+            .. note::
+
+               Throughput is calculated as:
+
+               - .. math:: throughput\_tot = requests \times (\mathsf{\text{input lengths}} + \mathsf{\text{output lengths}}) / elapsed\_time
+
+               - .. math:: throughput\_gen = requests \times \mathsf{\text{output lengths}} / elapsed\_time
+      {% endfor %}
+   {% endfor %}
+
+Further reading
+===============
+
+- To learn more about the options for latency and throughput benchmark scripts,
+  see `<https://github.com/ROCm/vllm/tree/main/benchmarks>`_.
+
+- To learn more about system settings and management practices to configure your system for
+  MI300X accelerators, see `AMD Instinct MI300X system optimization <https://instinct.docs.amd.com/projects/amdgpu-docs/en/latest/system-optimization/mi300x.html>`_
+
+- For application performance optimization strategies for HPC and AI workloads,
+  including inference with vLLM, see :doc:`/how-to/rocm-for-ai/inference-optimization/workload`.
+
+- To learn how to run community models from Hugging Face on AMD GPUs, see
+  :doc:`Running models from Hugging Face </how-to/rocm-for-ai/inference/hugging-face-models>`.
+
+- To learn how to fine-tune LLMs and optimize inference, see
+  :doc:`Fine-tuning LLMs and inference optimization </how-to/rocm-for-ai/fine-tuning/fine-tuning-and-inference>`.
+
+- For a list of other ready-made Docker images for AI with ROCm, see
+  `AMD Infinity Hub <https://www.amd.com/en/developer/resources/infinity-hub.html#f-amd_hub_category=AI%20%26%20ML%20Models>`_.
+
+Previous versions
+=================
+
+See :doc:`vllm-history` to find documentation for previous releases
+of the ``ROCm/vllm`` Docker image.

docs/how-to/rocm-for-ai/inference/benchmark-docker/previous-versions/vllm-0.9.1-20250702.rst

@@ -0,0 +1,353 @@
+:orphan:
+
+.. meta::
+   :description: Learn how to validate LLM inference performance on MI300X accelerators using AMD MAD and the
+                 ROCm vLLM Docker image.
+   :keywords: model, MAD, automation, dashboarding, validate
+
+**********************************
+vLLM inference performance testing
+**********************************
+
+.. caution::
+
+   This documentation does not reflect the latest version of ROCm vLLM
+   inference performance documentation. See :doc:`../vllm` for the latest version.
+
+.. _vllm-benchmark-unified-docker:
+
+.. datatemplate:yaml:: /data/how-to/rocm-for-ai/inference/previous-versions/vllm_0.9.1_20250702-benchmark-models.yaml
+
+   {% set unified_docker = data.vllm_benchmark.unified_docker.latest %}
+   {% set model_groups = data.vllm_benchmark.model_groups %}
+
+   The `ROCm vLLM Docker <{{ unified_docker.docker_hub_url }}>`_ image offers
+   a prebuilt, optimized environment for validating large language model (LLM)
+   inference performance on AMD Instinct™ MI300X series accelerators. This ROCm vLLM
+   Docker image integrates vLLM and PyTorch tailored specifically for MI300X series
+   accelerators and includes the following components:
+
+   * `ROCm {{ unified_docker.rocm_version }} <https://github.com/ROCm/ROCm>`_
+
+   * `vLLM {{ unified_docker.vllm_version }} <https://docs.vllm.ai/en/latest>`_
+
+   * `PyTorch {{ unified_docker.pytorch_version }} <https://github.com/ROCm/pytorch.git>`_
+
+   * `hipBLASLt {{ unified_docker.hipblaslt_version }} <https://github.com/ROCm/hipBLASLt>`_
+
+   With this Docker image, you can quickly test the :ref:`expected
+   inference performance numbers <vllm-benchmark-performance-measurements-20250702>` for
+   MI300X series accelerators.
+
+   .. _vllm-benchmark-available-models-20250702:
+
+   Supported models
+   ================
+
+   The following models are supported for inference performance benchmarking
+   with vLLM and ROCm. Some instructions, commands, and recommendations in this
+   documentation might vary by model -- select one to get started.
+
+   .. raw:: html
+
+      <div id="vllm-benchmark-ud-params-picker" class="container-fluid">
+        <div class="row">
+          <div class="col-2 me-2 model-param-head">Model group</div>
+          <div class="row col-10">
+   {% for model_group in model_groups %}
+            <div class="col-3 model-param" data-param-k="model-group" data-param-v="{{ model_group.tag }}" tabindex="0">{{ model_group.group }}</div>
+   {% endfor %}
+          </div>
+        </div>
+
+        <div class="row mt-1">
+          <div class="col-2 me-2 model-param-head">Model</div>
+          <div class="row col-10">
+   {% for model_group in model_groups %}
+      {% set models = model_group.models %}
+      {% for model in models %}
+         {% if models|length % 3 == 0 %}
+            <div class="col-4 model-param" data-param-k="model" data-param-v="{{ model.mad_tag }}" data-param-group="{{ model_group.tag }}" tabindex="0">{{ model.model }}</div>
+         {% else %}
+            <div class="col-6 model-param" data-param-k="model" data-param-v="{{ model.mad_tag }}" data-param-group="{{ model_group.tag }}" tabindex="0">{{ model.model }}</div>
+         {% endif %}
+      {% endfor %}
+   {% endfor %}
+          </div>
+        </div>
+      </div>
+
+   .. _vllm-benchmark-vllm:
+
+   {% for model_group in model_groups %}
+      {% for model in model_group.models %}
+
+   .. container:: model-doc {{model.mad_tag}}
+
+      .. note::
+
+         See the `{{ model.model }} model card on Hugging Face <{{ model.url }}>`_ to learn more about your selected model.
+         Some models require access authorization prior to use via an external license agreement through a third party.
+
+      {% endfor %}
+   {% endfor %}
+
+   .. note::
+
+      vLLM is a toolkit and library for LLM inference and serving. AMD implements
+      high-performance custom kernels and modules in vLLM to enhance performance.
+      See :ref:`fine-tuning-llms-vllm` and :ref:`mi300x-vllm-optimization` for
+      more information.
+
+   .. _vllm-benchmark-performance-measurements-20250702:
+
+   Performance measurements
+   ========================
+
+   To evaluate performance, the
+   `Performance results with AMD ROCm software <https://www.amd.com/en/developer/resources/rocm-hub/dev-ai/performance-results.html>`_
+   page provides reference throughput and latency measurements for inferencing popular AI models.
+
+   .. important::
+
+      The performance data presented in
+      `Performance results with AMD ROCm software <https://www.amd.com/en/developer/resources/rocm-hub/dev-ai/performance-results.html>`_
+      only reflects the latest version of this inference benchmarking environment.
+      The listed measurements should not be interpreted as the peak performance achievable by AMD Instinct MI325X and MI300X accelerators or ROCm software.
+
+   Advanced features and known issues
+   ==================================
+
+   For information on experimental features and known issues related to ROCm optimization efforts on vLLM,
+   see the developer's guide at `<https://github.com/ROCm/vllm/tree/5486e7bc8523be0324ccd68f221959445b56cc2a/docs/dev-docker>`__.
+
+   System validation
+   =================
+
+   Before running AI workloads, it's important to validate that your AMD hardware is configured
+   correctly and performing optimally.
+
+   To optimize performance, disable automatic NUMA balancing. Otherwise, the GPU
+   might hang until the periodic balancing is finalized. For more information,
+   see the :ref:`system validation steps <rocm-for-ai-system-optimization>`.
+
+   .. code-block:: shell
+
+      # disable automatic NUMA balancing
+      sh -c 'echo 0 > /proc/sys/kernel/numa_balancing'
+      # check if NUMA balancing is disabled (returns 0 if disabled)
+      cat /proc/sys/kernel/numa_balancing
+      0
+
+   To test for optimal performance, consult the recommended :ref:`System health benchmarks
+   <rocm-for-ai-system-health-bench>`. This suite of tests will help you verify and fine-tune your
+   system's configuration.
+
+   Pull the Docker image
+   =====================
+
+   Download the `ROCm vLLM Docker image <{{ unified_docker.docker_hub_url }}>`_.
+   Use the following command to pull the Docker image from Docker Hub.
+
+   .. code-block:: shell
+
+      docker pull {{ unified_docker.pull_tag }}
+
+   Benchmarking
+   ============
+
+   Once the setup is complete, choose between two options to reproduce the
+   benchmark results:
+
+   .. _vllm-benchmark-mad:
+
+   {% for model_group in model_groups %}
+      {% for model in model_group.models %}
+
+   .. container:: model-doc {{model.mad_tag}}
+
+      .. tab-set::
+
+         .. tab-item:: MAD-integrated benchmarking
+
+            Clone the ROCm Model Automation and Dashboarding (`<https://github.com/ROCm/MAD>`__) repository to a local
+            directory and install the required packages on the host machine.
+
+            .. code-block:: shell
+
+               git clone https://github.com/ROCm/MAD
+               cd MAD
+               pip install -r requirements.txt
+
+            Use this command to run the performance benchmark test on the `{{model.model}} <{{ model.url }}>`_ model
+            using one GPU with the :literal:`{{model.precision}}` data type on the host machine.
+
+            .. code-block:: shell
+
+               export MAD_SECRETS_HFTOKEN="your personal Hugging Face token to access gated models"
+               python3 tools/run_models.py --tags {{model.mad_tag}} --keep-model-dir --live-output --timeout 28800
+
+            MAD launches a Docker container with the name
+            ``container_ci-{{model.mad_tag}}``. The latency and throughput reports of the
+            model are collected in the following path: ``~/MAD/reports_{{model.precision}}/``.
+
+            Although the :ref:`available models <vllm-benchmark-available-models-20250702>` are preconfigured
+            to collect latency and throughput performance data, you can also change the benchmarking
+            parameters. See the standalone benchmarking tab for more information.
+
+            {% if model.tunableop %}
+
+            .. note::
+
+               For improved performance, consider enabling :ref:`PyTorch TunableOp <mi300x-tunableop>`.
+               TunableOp automatically explores different implementations and configurations of certain PyTorch
+               operators to find the fastest one for your hardware.
+
+               By default, ``{{model.mad_tag}}`` runs with TunableOp disabled
+               (see
+               `<https://github.com/ROCm/MAD/blob/develop/models.json>`__). To
+               enable it, edit the default run behavior in the ``models.json``
+               configuration before running inference -- update the model's run
+               ``args`` by changing ``--tunableop off`` to ``--tunableop on``.
+
+               Enabling TunableOp triggers a two-pass run -- a warm-up followed by the performance-collection run.
+
+            {% endif %}
+
+         .. tab-item:: Standalone benchmarking
+
+            Run the vLLM benchmark tool independently by starting the
+            `Docker container <{{ unified_docker.docker_hub_url }}>`_
+            as shown in the following snippet.
+
+            .. code-block::
+
+               docker pull {{ unified_docker.pull_tag }}
+               docker run -it --device=/dev/kfd --device=/dev/dri --group-add video --shm-size 16G --security-opt seccomp=unconfined --security-opt apparmor=unconfined --cap-add=SYS_PTRACE -v $(pwd):/workspace --env HUGGINGFACE_HUB_CACHE=/workspace --name test {{ unified_docker.pull_tag }}
+
+            In the Docker container, clone the ROCm MAD repository and navigate to the
+            benchmark scripts directory at ``~/MAD/scripts/vllm``.
+
+            .. code-block::
+
+               git clone https://github.com/ROCm/MAD
+               cd MAD/scripts/vllm
+
+            To start the benchmark, use the following command with the appropriate options.
+
+            .. code-block::
+
+               ./vllm_benchmark_report.sh -s $test_option -m {{model.model_repo}} -g $num_gpu -d {{model.precision}}
+
+            .. list-table::
+               :header-rows: 1
+               :align: center
+
+               * - Name
+                 - Options
+                 - Description
+
+               * - ``$test_option``
+                 - latency
+                 - Measure decoding token latency
+
+               * -
+                 - throughput
+                 - Measure token generation throughput
+
+               * -
+                 - all
+                 - Measure both throughput and latency
+
+               * - ``$num_gpu``
+                 - 1 or 8
+                 - Number of GPUs
+
+               * - ``$datatype``
+                 - ``float16`` or ``float8``
+                 - Data type
+
+            .. note::
+
+               The input sequence length, output sequence length, and tensor parallel (TP) are
+               already configured. You don't need to specify them with this script.
+
+            .. note::
+
+               If you encounter the following error, pass your access-authorized Hugging
+               Face token to the gated models.
+
+               .. code-block::
+
+                  OSError: You are trying to access a gated repo.
+
+                  # pass your HF_TOKEN
+                  export HF_TOKEN=$your_personal_hf_token
+
+            Here are some examples of running the benchmark with various options.
+
+            * Latency benchmark
+
+              Use this command to benchmark the latency of the {{model.model}} model on eight GPUs with :literal`{{model.precision}}` precision.
+
+              .. code-block::
+
+                 ./vllm_benchmark_report.sh -s latency -m {{model.model_repo}} -g 8 -d {{model.precision}}
+
+              Find the latency report at ``./reports_{{model.precision}}_vllm_rocm{{unified_docker.rocm_version}}/summary/{{model.model_repo.split('/', 1)[1] if '/' in model.model_repo else model.model_repo}}_latency_report.csv``.
+
+            * Throughput benchmark
+
+              Use this command to benchmark the throughput of the {{model.model}} model on eight GPUs with :literal:`{{model.precision}}` precision.
+
+              .. code-block:: shell
+
+                 ./vllm_benchmark_report.sh -s throughput -m {{model.model_repo}} -g 8 -d {{model.precision}}
+
+              Find the throughput report at ``./reports_{{model.precision}}_vllm_rocm{{unified_docker.rocm_version}}/summary/{{model.model_repo.split('/', 1)[1] if '/' in model.model_repo else model.model_repo}}_throughput_report.csv``.
+
+            .. raw:: html
+
+               <style>
+               mjx-container[jax="CHTML"][display="true"] {
+                  text-align: left;
+                  margin: 0;
+               }
+               </style>
+
+            .. note::
+
+               Throughput is calculated as:
+
+               - .. math:: throughput\_tot = requests \times (\mathsf{\text{input lengths}} + \mathsf{\text{output lengths}}) / elapsed\_time
+
+               - .. math:: throughput\_gen = requests \times \mathsf{\text{output lengths}} / elapsed\_time
+      {% endfor %}
+   {% endfor %}
+
+Further reading
+===============
+
+- To learn more about the options for latency and throughput benchmark scripts,
+  see `<https://github.com/ROCm/vllm/tree/main/benchmarks>`_.
+
+- To learn more about system settings and management practices to configure your system for
+  MI300X series accelerators, see `AMD Instinct MI300X system optimization <https://instinct.docs.amd.com/projects/amdgpu-docs/en/latest/system-optimization/mi300x.html>`_
+
+- For application performance optimization strategies for HPC and AI workloads,
+  including inference with vLLM, see :doc:`/how-to/rocm-for-ai/inference-optimization/workload`.
+
+- To learn how to run community models from Hugging Face on AMD GPUs, see
+  :doc:`Running models from Hugging Face </how-to/rocm-for-ai/inference/hugging-face-models>`.
+
+- To learn how to fine-tune LLMs and optimize inference, see
+  :doc:`Fine-tuning LLMs and inference optimization </how-to/rocm-for-ai/fine-tuning/fine-tuning-and-inference>`.
+
+- For a list of other ready-made Docker images for AI with ROCm, see
+  `AMD Infinity Hub <https://www.amd.com/en/developer/resources/infinity-hub.html#f-amd_hub_category=AI%20%26%20ML%20Models>`_.
+
+Previous versions
+=================
+
+See :doc:`vllm-history` to find documentation for previous releases
+of the ``ROCm/vllm`` Docker image.

docs/how-to/rocm-for-ai/inference/benchmark-docker/previous-versions/vllm-history.rst

@@ -7,69 +7,103 @@ vLLM inference performance testing version history
 This table lists previous versions of the ROCm vLLM inference Docker image for
 inference performance testing. For detailed information about available models
 for benchmarking, see the version-specific documentation. You can find tagged
-previous releases of the ``ROCm/vllm`` Docker image on `Docker Hub <https://hub.docker.com/r/rocm/vllm/tags>`_.
+previous releases of the ``ROCm/vllm`` Docker image on `Docker Hub <https://hub.docker.com/r/rocm/vllm/tags>`__.
 
 .. list-table::
    :header-rows: 1
-   :stub-columns: 1
 
-   * - ROCm version
+   * - Docker image tag
-     - vLLM version
+     - Components
-     - PyTorch version
      - Resources
 
-   * - 6.4.0
+   * - ``rocm/vllm:rocm6.4.1_vllm_0.9.1_20250715``
-     - 0.9.0.1
+       (latest)
-     - 2.7.0
+     - 
+       * ROCm 6.4.1
+       * vLLM 0.9.1
+       * PyTorch 2.7.0
      - 
        * :doc:`Documentation <../vllm>`
-       * `Docker Hub <https://hub.docker.com/layers/rocm/vllm/rocm6.3.1_vllm_0.8.5_20250521/images/sha256-38410c51af7208897cd8b737c9bdfc126e9bc8952d4aa6b88c85482f03092a11>`_
+       * `Docker Hub <https://hub.docker.com/layers/rocm/vllm/rocm6.4.1_vllm_0.9.1_20250715/images/sha256-4a429705fa95a58f6d20aceab43b1b76fa769d57f32d5d28bd3f4e030e2a78ea>`__
 
-   * - 6.3.1
+   * - ``rocm/vllm:rocm6.4.1_vllm_0.9.1_20250702``
-     - 0.8.5 (0.8.6.dev)
+     - 
-     - 2.7.0
+       * ROCm 6.4.1
+       * vLLM 0.9.1
+       * PyTorch 2.7.0
+     - 
+       * :doc:`Documentation <vllm-0.9.1-20250702>`
+       * `Docker Hub <https://hub.docker.com/layers/rocm/vllm/rocm6.4.1_vllm_0.9.1_20250702/images/sha256-45068a2079cb8df554ed777141bf0c67d6627c470a897256e60c9f262677faab>`__
+
+   * - ``rocm/vllm:rocm6.4.1_vllm_0.9.0.1_20250605``
+     - 
+       * ROCm 6.4.1
+       * vLLM 0.9.0.1
+       * PyTorch 2.7.0
+     - 
+       * :doc:`Documentation <vllm-0.9.0.1-20250605>`
+       * `Docker Hub <https://hub.docker.com/layers/rocm/vllm/rocm6.4.1_vllm_0.9.0.1_20250605/images/sha256-f48beeb3d72663a93c77211eb45273d564451447c097e060befa713d565fa36c>`__
+
+   * - ``rocm/vllm:rocm6.3.1_vllm_0.8.5_20250521``
+     - 
+       * ROCm 6.3.1
+       * 0.8.5 vLLM (0.8.6.dev)
+       * PyTorch 2.7.0
      - 
        * :doc:`Documentation <vllm-0.8.5-20250521>`
-       * `Docker Hub <https://hub.docker.com/layers/rocm/vllm/rocm6.3.1_vllm_0.8.5_20250521/images/sha256-38410c51af7208897cd8b737c9bdfc126e9bc8952d4aa6b88c85482f03092a11>`_
+       * `Docker Hub <https://hub.docker.com/layers/rocm/vllm/rocm6.3.1_vllm_0.8.5_20250521/images/sha256-38410c51af7208897cd8b737c9bdfc126e9bc8952d4aa6b88c85482f03092a11>`__
 
-   * - 6.3.1
+   * - ``rocm/vllm:rocm6.3.1_vllm_0.8.5_20250513``
-     - 0.8.5
+     - 
-     - 2.7.0
+       * ROCm 6.3.1
+       * vLLM 0.8.5
+       * PyTorch 2.7.0
      - 
        * :doc:`Documentation <vllm-0.8.5-20250513>`
-       * `Docker Hub <https://hub.docker.com/layers/rocm/vllm/rocm6.3.1_vllm_0.8.5_20250513/images/sha256-5c8b4436dd0464119d9df2b44c745fadf81512f18ffb2f4b5dc235c71ebe26b4>`_
+       * `Docker Hub <https://hub.docker.com/layers/rocm/vllm/rocm6.3.1_vllm_0.8.5_20250513/images/sha256-5c8b4436dd0464119d9df2b44c745fadf81512f18ffb2f4b5dc235c71ebe26b4>`__
 
-   * - 6.3.1
+   * - ``rocm/vllm:rocm6.3.1_instinct_vllm0.8.3_20250415``
-     - 0.8.3
+     - 
-     - 2.7.0
+       * ROCm 6.3.1
+       * vLLM 0.8.3
+       * PyTorch 2.7.0
      - 
        * :doc:`Documentation <vllm-0.8.3-20250415>`
-       * `Docker Hub <https://hub.docker.com/layers/rocm/vllm/rocm6.3.1_instinct_vllm0.8.3_20250415/images/sha256-ad9062dea3483d59dedb17c67f7c49f30eebd6eb37c3fac0a171fb19696cc845>`_
+       * `Docker Hub <https://hub.docker.com/layers/rocm/vllm/rocm6.3.1_instinct_vllm0.8.3_20250415/images/sha256-ad9062dea3483d59dedb17c67f7c49f30eebd6eb37c3fac0a171fb19696cc845>`__
 
-   * - 6.3.1
+   * - ``rocm/vllm:rocm6.3.1_instinct_vllm0.7.3_20250325``
-     - 0.7.3
+     - 
-     - 2.7.0
+       * ROCm 6.3.1
+       * vLLM 0.7.3
+       * PyTorch 2.7.0
      - 
        * :doc:`Documentation <vllm-0.7.3-20250325>`
-       * `Docker Hub <https://hub.docker.com/layers/rocm/vllm/rocm6.3.1_instinct_vllm0.7.3_20250325/images/sha256-25245924f61750b19be6dcd8e787e46088a496c1fe17ee9b9e397f3d84d35640>`_
+       * `Docker Hub <https://hub.docker.com/layers/rocm/vllm/rocm6.3.1_instinct_vllm0.7.3_20250325/images/sha256-25245924f61750b19be6dcd8e787e46088a496c1fe17ee9b9e397f3d84d35640>`__
 
-   * - 6.3.1
+   * - ``rocm/vllm:rocm6.3.1_mi300_ubuntu22.04_py3.12_vllm_0.6.6``
-     - 0.6.6
+     - 
-     - 2.7.0
+       * ROCm 6.3.1
+       * vLLM 0.6.6
+       * PyTorch 2.7.0
      - 
        * :doc:`Documentation <vllm-0.6.6>`
-       * `Docker Hub <https://hub.docker.com/layers/rocm/vllm/rocm6.3.1_mi300_ubuntu22.04_py3.12_vllm_0.6.6/images/sha256-9a12ef62bbbeb5a4c30a01f702c8e025061f575aa129f291a49fbd02d6b4d6c9>`_
+       * `Docker Hub <https://hub.docker.com/layers/rocm/vllm/rocm6.3.1_mi300_ubuntu22.04_py3.12_vllm_0.6.6/images/sha256-9a12ef62bbbeb5a4c30a01f702c8e025061f575aa129f291a49fbd02d6b4d6c9>`__
 
-   * - 6.2.1
+   * - ``rocm/vllm:rocm6.2_mi300_ubuntu20.04_py3.9_vllm_0.6.4``
-     - 0.6.4
+     - 
-     - 2.5.0
+       * ROCm 6.2.1
+       * vLLM 0.6.4
+       * PyTorch 2.5.0
      - 
        * :doc:`Documentation <vllm-0.6.4>`
-       * `Docker Hub <https://hub.docker.com/layers/rocm/vllm/rocm6.2_mi300_ubuntu20.04_py3.9_vllm_0.6.4/images/sha256-ccbb74cc9e7adecb8f7bdab9555f7ac6fc73adb580836c2a35ca96ff471890d8>`_
+       * `Docker Hub <https://hub.docker.com/layers/rocm/vllm/rocm6.2_mi300_ubuntu20.04_py3.9_vllm_0.6.4/images/sha256-ccbb74cc9e7adecb8f7bdab9555f7ac6fc73adb580836c2a35ca96ff471890d8>`__
 
-   * - 6.2.0
+   * - ``rocm/vllm:rocm6.2_mi300_ubuntu22.04_py3.9_vllm_7c5fd50``
-     - 0.4.3
+     - 
-     - 2.4.0
+       * ROCm 6.2.0
+       * vLLM 0.4.3
+       * PyTorch 2.4.0
      - 
        * :doc:`Documentation <vllm-0.4.3>`
-       * `Docker Hub <https://hub.docker.com/layers/rocm/vllm/rocm6.2_mi300_ubuntu22.04_py3.9_vllm_7c5fd50/images/sha256-9e4dd4788a794c3d346d7d0ba452ae5e92d39b8dfac438b2af8efdc7f15d22c0>`_
+       * `Docker Hub <https://hub.docker.com/layers/rocm/vllm/rocm6.2_mi300_ubuntu22.04_py3.9_vllm_7c5fd50/images/sha256-9e4dd4788a794c3d346d7d0ba452ae5e92d39b8dfac438b2af8efdc7f15d22c0>`__
+

docs/how-to/rocm-for-ai/inference/benchmark-docker/pytorch-inference.rst

@@ -32,10 +32,10 @@ PyTorch inference performance testing
 
       <div id="vllm-benchmark-ud-params-picker" class="container-fluid">
         <div class="row">
-          <div class="col-2 me-2 model-param-head">Model group</div>
+          <div class="col-2 me-2 model-param-head">Model</div>
           <div class="row col-10">
    {% for model_group in model_groups %}
-            <div class="col-4 model-param" data-param-k="model-group" data-param-v="{{ model_group.tag }}" tabindex="0">{{ model_group.group }}</div>
+            <div class="col-3 model-param" data-param-k="model-group" data-param-v="{{ model_group.tag }}" tabindex="0">{{ model_group.group }}</div>
    {% endfor %}
           </div>
         </div>
@@ -93,7 +93,7 @@ PyTorch inference performance testing
 
    .. container:: model-doc pyt_chai1_inference
 
-      Use the following command to pull the `ROCm PyTorch Docker image <https://hub.docker.com/layers/rocm/pytorch/rocm6.2.3_ubuntu22.04_py3.10_pytorch_release_2.3.0_triton_llvm_reg_issue/images/sha256-b736a4239ab38a9d0e448af6d4adca83b117debed00bfbe33846f99c4540f79b>`_ from Docker Hub.
+      Use the following command to pull the `ROCm PyTorch Docker image <https://hub.docker.com/layers/rocm/pytorch/rocm6.2.3_ubuntu22.04_py3.10_pytorch_release_2.3.0_triton_llvm_reg_issue/images/sha256-b736a4239ab38a9d0e448af6d4adca83b117debed00bfbe33846f99c4540f79b>`__ from Docker Hub.
 
       .. code-block:: shell
 
@@ -103,9 +103,9 @@ PyTorch inference performance testing
 
          The Chai-1 benchmark uses a specifically selected Docker image using ROCm 6.2.3 and PyTorch 2.3.0 to address an accuracy issue.
 
-   .. container:: model-doc pyt_clip_inference pyt_mochi_video_inference
+   .. container:: model-doc pyt_clip_inference pyt_mochi_video_inference pyt_wan2.1_inference
 
-      Use the following command to pull the `ROCm PyTorch Docker image <https://hub.docker.com/layers/rocm/pytorch/latest/images/sha256-05b55983e5154f46e7441897d0908d79877370adca4d1fff4899d9539d6c4969>`_ from Docker Hub.
+      Use the following command to pull the `ROCm PyTorch Docker image <https://hub.docker.com/layers/rocm/pytorch/latest/images/sha256-05b55983e5154f46e7441897d0908d79877370adca4d1fff4899d9539d6c4969>`__ from Docker Hub.
 
       .. code-block:: shell
 

docs/how-to/rocm-for-ai/inference/benchmark-docker/vllm.rst

@@ -20,23 +20,55 @@ vLLM inference performance testing
    Docker image integrates vLLM and PyTorch tailored specifically for MI300X series
    accelerators and includes the following components:
 
-   * `ROCm {{ unified_docker.rocm_version }} <https://github.com/ROCm/ROCm>`_
+   .. list-table::
+      :header-rows: 1
 
-   * `vLLM {{ unified_docker.vllm_version }} <https://docs.vllm.ai/en/latest>`_
+      * - Software component
+        - Version
 
-   * `PyTorch {{ unified_docker.pytorch_version }} <https://github.com/ROCm/pytorch.git>`_
+      * - `ROCm <https://github.com/ROCm/ROCm>`__
+        - {{ unified_docker.rocm_version }}
 
-   * `hipBLASLt {{ unified_docker.hipblaslt_version }} <https://github.com/ROCm/hipBLASLt>`_
+      * - `vLLM <https://docs.vllm.ai/en/latest>`__
+        - {{ unified_docker.vllm_version }}
 
-   With this Docker image, you can quickly test the :ref:`expected
+      * - `PyTorch <https://github.com/ROCm/pytorch>`__
-   inference performance numbers <vllm-benchmark-performance-measurements>` for
+        - {{ unified_docker.pytorch_version }}
-   MI300X series accelerators.
+
+      * - `hipBLASLt <https://github.com/ROCm/hipBLASLt>`__
+        - {{ unified_docker.hipblaslt_version }}
+
+With this Docker image, you can quickly test the :ref:`expected
+inference performance numbers <vllm-benchmark-performance-measurements>` for
+MI300X series accelerators.
+
+What's new
+==========
+
+The following is summary of notable changes since the :doc:`previous ROCm/vLLM Docker release <previous-versions/vllm-history>`.
+
+* The ``--compilation-config-parameter`` is no longer required as its options are now enabled by default.
+  This parameter has been removed from the benchmarking script.
+
+* Resolved Llama 3.1 405 B custom all-reduce issue, eliminating the need for ``--disable-custom-all-reduce``.
+  This parameter has been removed from the benchmarking script.
+
+* Fixed a ``+rms_norm`` custom kernel issue.
+
+* Added quick reduce functionality. Set ``VLLM_ROCM_QUICK_REDUCE_QUANTIZATION=FP`` to enable; supported modes are ``FP``, ``INT8``, ``INT6``, ``INT4``.
+
+* Implemented a workaround to potentially mitigate GPU crashes experienced with the Command R+ model, pending a driver fix.
+
+Supported models
+================
+
+.. datatemplate:yaml:: /data/how-to/rocm-for-ai/inference/vllm-benchmark-models.yaml
+
+   {% set unified_docker = data.vllm_benchmark.unified_docker.latest %}
+   {% set model_groups = data.vllm_benchmark.model_groups %}
 
    .. _vllm-benchmark-available-models:
 
-   Supported models
-   ================
-
    The following models are supported for inference performance benchmarking
    with vLLM and ROCm. Some instructions, commands, and recommendations in this
    documentation might vary by model -- select one to get started.
@@ -44,18 +76,18 @@ vLLM inference performance testing
    .. raw:: html
 
       <div id="vllm-benchmark-ud-params-picker" class="container-fluid">
-        <div class="row">
+      <div class="row">
-          <div class="col-2 me-2 model-param-head">Model group</div>
+         <div class="col-2 me-2 model-param-head">Model group</div>
-          <div class="row col-10">
+         <div class="row col-10">
    {% for model_group in model_groups %}
             <div class="col-3 model-param" data-param-k="model-group" data-param-v="{{ model_group.tag }}" tabindex="0">{{ model_group.group }}</div>
    {% endfor %}
-          </div>
+         </div>
-        </div>
+      </div>
 
-        <div class="row mt-1">
+      <div class="row mt-1">
-          <div class="col-2 me-2 model-param-head">Model</div>
+         <div class="col-2 me-2 model-param-head">Model</div>
-          <div class="row col-10">
+         <div class="row col-10">
    {% for model_group in model_groups %}
       {% set models = model_group.models %}
       {% for model in models %}
@@ -66,8 +98,8 @@ vLLM inference performance testing
          {% endif %}
       {% endfor %}
    {% endfor %}
-          </div>
+         </div>
-        </div>
+      </div>
       </div>
 
    .. _vllm-benchmark-vllm:
@@ -85,56 +117,48 @@ vLLM inference performance testing
       {% endfor %}
    {% endfor %}
 
-   .. note::
+.. note::
 
-      vLLM is a toolkit and library for LLM inference and serving. AMD implements
+   vLLM is a toolkit and library for LLM inference and serving. AMD implements
-      high-performance custom kernels and modules in vLLM to enhance performance.
+   high-performance custom kernels and modules in vLLM to enhance performance.
-      See :ref:`fine-tuning-llms-vllm` and :ref:`mi300x-vllm-optimization` for
+   See :ref:`fine-tuning-llms-vllm` and :ref:`mi300x-vllm-optimization` for
-      more information.
+   more information.
 
-   .. _vllm-benchmark-performance-measurements:
+.. _vllm-benchmark-performance-measurements:
 
-   Performance measurements
+Performance measurements
-   ========================
+========================
 
-   To evaluate performance, the
+To evaluate performance, the
+`Performance results with AMD ROCm software <https://www.amd.com/en/developer/resources/rocm-hub/dev-ai/performance-results.html>`_
+page provides reference throughput and latency measurements for inferencing popular AI models.
+
+.. important::
+
+   The performance data presented in
    `Performance results with AMD ROCm software <https://www.amd.com/en/developer/resources/rocm-hub/dev-ai/performance-results.html>`_
-   page provides reference throughput and latency measurements for inferencing popular AI models.
+   only reflects the latest version of this inference benchmarking environment.
+   The listed measurements should not be interpreted as the peak performance achievable by AMD Instinct MI325X and MI300X accelerators or ROCm software.
 
-   .. important::
+System validation
+=================
 
-      The performance data presented in
+Before running AI workloads, it's important to validate that your AMD hardware is configured
-      `Performance results with AMD ROCm software <https://www.amd.com/en/developer/resources/rocm-hub/dev-ai/performance-results.html>`_
+correctly and performing optimally.
-      only reflects the latest version of this inference benchmarking environment.
-      The listed measurements should not be interpreted as the peak performance achievable by AMD Instinct MI325X and MI300X accelerators or ROCm software.
 
-   Advanced features and known issues
+If you have already validated your system settings, including aspects like NUMA auto-balancing, you
-   ==================================
+can skip this step. Otherwise, complete the procedures in the :ref:`System validation and
+optimization <rocm-for-ai-system-optimization>` guide to properly configure your system settings
+before starting training.
 
-   For information on experimental features and known issues related to ROCm optimization efforts on vLLM,
+To test for optimal performance, consult the recommended :ref:`System health benchmarks
-   see the developer's guide at `<https://github.com/ROCm/vllm/tree/7bb0618b1fe725b7d4fad9e525aa44da12c94a8b/docs/dev-docker>`__.
+<rocm-for-ai-system-health-bench>`. This suite of tests will help you verify and fine-tune your
+system's configuration.
 
-   System validation
+.. datatemplate:yaml:: /data/how-to/rocm-for-ai/inference/vllm-benchmark-models.yaml
-   =================
 
-   Before running AI workloads, it's important to validate that your AMD hardware is configured
+   {% set unified_docker = data.vllm_benchmark.unified_docker.latest %}
-   correctly and performing optimally.
+   {% set model_groups = data.vllm_benchmark.model_groups %}
-
-   To optimize performance, disable automatic NUMA balancing. Otherwise, the GPU
-   might hang until the periodic balancing is finalized. For more information,
-   see the :ref:`system validation steps <rocm-for-ai-system-optimization>`.
-
-   .. code-block:: shell
-
-      # disable automatic NUMA balancing
-      sh -c 'echo 0 > /proc/sys/kernel/numa_balancing'
-      # check if NUMA balancing is disabled (returns 0 if disabled)
-      cat /proc/sys/kernel/numa_balancing
-      0
-
-   To test for optimal performance, consult the recommended :ref:`System health benchmarks
-   <rocm-for-ai-system-health-bench>`. This suite of tests will help you verify and fine-tune your
-   system's configuration.
 
    Pull the Docker image
    =====================
@@ -163,22 +187,26 @@ vLLM inference performance testing
 
          .. tab-item:: MAD-integrated benchmarking
 
-            Clone the ROCm Model Automation and Dashboarding (`<https://github.com/ROCm/MAD>`__) repository to a local
+            1. Clone the ROCm Model Automation and Dashboarding (`<https://github.com/ROCm/MAD>`__) repository to a local
-            directory and install the required packages on the host machine.
+               directory and install the required packages on the host machine.
 
-            .. code-block:: shell
+               .. code-block:: shell
 
-               git clone https://github.com/ROCm/MAD
+                  git clone https://github.com/ROCm/MAD
-               cd MAD
+                  cd MAD
-               pip install -r requirements.txt
+                  pip install -r requirements.txt
 
-            Use this command to run the performance benchmark test on the `{{model.model}} <{{ model.url }}>`_ model
+            2. Use this command to run the performance benchmark test on the `{{model.model}} <{{ model.url }}>`_ model
-            using one GPU with the ``{{model.precision}}`` data type on the host machine.
+               using one GPU with the :literal:`{{model.precision}}` data type on the host machine.
 
-            .. code-block:: shell
+               .. code-block:: shell
 
-               export MAD_SECRETS_HFTOKEN="your personal Hugging Face token to access gated models"
+                  export MAD_SECRETS_HFTOKEN="your personal Hugging Face token to access gated models"
-               python3 tools/run_models.py --tags {{model.mad_tag}} --keep-model-dir --live-output --timeout 28800
+                  python3 tools/run_models.py \
+                      --tags {{model.mad_tag}} \
+                      --keep-model-dir \
+                      --live-output \
+                      --timeout 28800
 
             MAD launches a Docker container with the name
             ``container_ci-{{model.mad_tag}}``. The latency and throughput reports of the
@@ -209,93 +237,125 @@ vLLM inference performance testing
 
          .. tab-item:: Standalone benchmarking
 
-            Run the vLLM benchmark tool independently by starting the
+            .. rubric:: Download the Docker image and required scripts
-            `Docker container <{{ unified_docker.docker_hub_url }}>`_
-            as shown in the following snippet.
 
-            .. code-block::
+            1. Run the vLLM benchmark tool independently by starting the
+               `Docker container <{{ unified_docker.docker_hub_url }}>`_
+               as shown in the following snippet.
 
-               docker pull {{ unified_docker.pull_tag }}
+               .. code-block:: shell
-               docker run -it --device=/dev/kfd --device=/dev/dri --group-add video --shm-size 16G --security-opt seccomp=unconfined --security-opt apparmor=unconfined --cap-add=SYS_PTRACE -v $(pwd):/workspace --env HUGGINGFACE_HUB_CACHE=/workspace --name test {{ unified_docker.pull_tag }}
 
-            In the Docker container, clone the ROCm MAD repository and navigate to the
+                  docker pull {{ unified_docker.pull_tag }}
-            benchmark scripts directory at ``~/MAD/scripts/vllm``.
+                  docker run -it \
+                      --device=/dev/kfd \
+                      --device=/dev/dri \
+                      --group-add video \
+                      --shm-size 16G \
+                      --security-opt seccomp=unconfined \
+                      --security-opt apparmor=unconfined \
+                      --cap-add=SYS_PTRACE \
+                      -v $(pwd):/workspace \
+                      --env HUGGINGFACE_HUB_CACHE=/workspace \
+                      --name test \
+                      {{ unified_docker.pull_tag }}
 
-            .. code-block::
+            2. In the Docker container, clone the ROCm MAD repository and navigate to the
+               benchmark scripts directory at ``~/MAD/scripts/vllm``.
 
-               git clone https://github.com/ROCm/MAD
+               .. code-block:: shell
-               cd MAD/scripts/vllm
 
-            To start the benchmark, use the following command with the appropriate options.
+                  git clone https://github.com/ROCm/MAD
+                  cd MAD/scripts/vllm
 
-            .. code-block::
+            3. To start the benchmark, use the following command with the appropriate options.
 
-               ./vllm_benchmark_report.sh -s $test_option -m {{model.model_repo}} -g $num_gpu -d {{model.precision}}
+               .. dropdown:: Benchmark options
+                  :open:
 
-            .. list-table::
+                  .. list-table::
-               :header-rows: 1
+                     :header-rows: 1
-               :align: center
+                     :align: center
 
-               * - Name
+                     * - Name
-                 - Options
+                       - Options
-                 - Description
+                       - Description
 
-               * - ``$test_option``
+                     * - ``$test_option``
-                 - latency
+                       - latency
-                 - Measure decoding token latency
+                       - Measure decoding token latency
 
-               * -
+                     * -
-                 - throughput
+                       - throughput
-                 - Measure token generation throughput
+                       - Measure token generation throughput
 
-               * -
+                     * -
-                 - all
+                       - all
-                 - Measure both throughput and latency
+                       - Measure both throughput and latency
 
-               * - ``$num_gpu``
+                     * - ``$num_gpu``
-                 - 1 or 8
+                       - 1 or 8
-                 - Number of GPUs
+                       - Number of GPUs
 
-               * - ``$datatype``
+                     * - ``$datatype``
-                 - ``float16`` or ``float8``
+                       - ``float16`` or ``float8``
-                 - Data type
+                       - Data type
 
-            .. note::
+                  The input sequence length, output sequence length, and tensor parallel (TP) are
+                  already configured. You don't need to specify them with this script.
 
-               The input sequence length, output sequence length, and tensor parallel (TP) are
+               Command:
-               already configured. You don't need to specify them with this script.
-
-            .. note::
-
-               If you encounter the following error, pass your access-authorized Hugging
-               Face token to the gated models.
 
                .. code-block::
 
-                  OSError: You are trying to access a gated repo.
+                  ./vllm_benchmark_report.sh \
+                      -s $test_option \
+                      -m {{model.model_repo}} \
+                      -g $num_gpu \
+                      -d {{model.precision}}
 
-                  # pass your HF_TOKEN
+               .. note::
-                  export HF_TOKEN=$your_personal_hf_token
 
-            Here are some examples of running the benchmark with various options.
+                  For best performance, it's recommend to run with ``VLLM_V1_USE_PREFILL_DECODE_ATTENTION=1``.
+
+                  If you encounter the following error, pass your access-authorized Hugging
+                  Face token to the gated models.
+
+                  .. code-block::
+
+                     OSError: You are trying to access a gated repo.
+
+                     # pass your HF_TOKEN
+                     export HF_TOKEN=$your_personal_hf_token
+
+            .. rubric:: Benchmarking examples
+
+            Here are some examples of running the benchmark with various options:
 
             * Latency benchmark
 
-              Use this command to benchmark the latency of the {{model.model}} model on eight GPUs with ``{{model.precision}}`` precision.
+              Use this command to benchmark the latency of the {{model.model}} model on eight GPUs with :literal:`{{model.precision}}` precision.
 
               .. code-block::
 
-                 ./vllm_benchmark_report.sh -s latency -m {{model.model_repo}} -g 8 -d {{model.precision}}
+                 ./vllm_benchmark_report.sh \
+                     -s latency \
+                     -m {{model.model_repo}} \
+                     -g 8 \
+                     -d {{model.precision}}
 
               Find the latency report at ``./reports_{{model.precision}}_vllm_rocm{{unified_docker.rocm_version}}/summary/{{model.model_repo.split('/', 1)[1] if '/' in model.model_repo else model.model_repo}}_latency_report.csv``.
 
             * Throughput benchmark
 
-              Use this command to benchmark the throughput of the {{model.model}} model on eight GPUs with ``{{model.precision}}`` precision.
+              Use this command to benchmark the throughput of the {{model.model}} model on eight GPUs with :literal:`{{model.precision}}` precision.
 
               .. code-block:: shell
 
-                 ./vllm_benchmark_report.sh -s throughput -m {{model.model_repo}} -g 8 -d {{model.precision}}
+                 ./vllm_benchmark_report.sh \
+                     -s throughput \
+                     -m {{model.model_repo}} \
+                     -g 8 \
+                     -d {{model.precision}}
 
               Find the throughput report at ``./reports_{{model.precision}}_vllm_rocm{{unified_docker.rocm_version}}/summary/{{model.model_repo.split('/', 1)[1] if '/' in model.model_repo else model.model_repo}}_throughput_report.csv``.
 
@@ -318,6 +378,41 @@ vLLM inference performance testing
       {% endfor %}
    {% endfor %}
 
+Advanced usage
+==============
+
+For information on experimental features and known issues related to ROCm optimization efforts on vLLM,
+see the developer's guide at `<https://github.com/ROCm/vllm/tree/f94ec9beeca1071cc34f9d1e206d8c7f3ac76129/docs/dev-docker>`__.
+
+Reproducing the Docker image
+----------------------------
+
+To reproduce this ROCm/vLLM Docker image release, follow these steps:
+
+1. Clone the `vLLM repository <https://github.com/ROCm/vllm>`__.
+
+   .. code-block:: shell
+
+      git clone https://github.com/ROCm/vllm.git
+
+2. Checkout the specific release commit.
+
+   .. code-block:: shell
+
+      cd vllm
+      git checkout b432b7a285aa0dcb9677380936ffa74931bb6d6f
+
+3. Build the Docker image. Replace ``vllm-rocm`` with your desired image tag.
+
+   .. code-block:: shell
+
+      docker build -f docker/Dockerfile.rocm -t vllm-rocm .
+
+Known issues and workarounds
+============================
+
+AITER does not support FP8 KV cache yet.
+
 Further reading
 ===============
 
@@ -325,22 +420,22 @@ Further reading
   see `<https://github.com/ROCm/vllm/tree/main/benchmarks>`_.
 
 - To learn more about system settings and management practices to configure your system for
-  MI300X accelerators, see `AMD Instinct MI300X system optimization <https://instinct.docs.amd.com/projects/amdgpu-docs/en/latest/system-optimization/mi300x.html>`_
+  MI300X series accelerators, see `AMD Instinct MI300X system optimization <https://instinct.docs.amd.com/projects/amdgpu-docs/en/latest/system-optimization/mi300x.html>`_
 
 - For application performance optimization strategies for HPC and AI workloads,
-  including inference with vLLM, see :doc:`../../inference-optimization/workload`.
+  including inference with vLLM, see :doc:`/how-to/rocm-for-ai/inference-optimization/workload`.
 
-- To learn how to run LLM models from Hugging Face or your own model, see
+- To learn how to run community models from Hugging Face on AMD GPUs, see
-  :doc:`Running models from Hugging Face <../hugging-face-models>`.
+  :doc:`Running models from Hugging Face </how-to/rocm-for-ai/inference/hugging-face-models>`.
 
-- To learn how to optimize inference on LLMs, see
+- To learn how to fine-tune LLMs and optimize inference, see
-  :doc:`Inference optimization <../../inference-optimization/index>`.
+  :doc:`Fine-tuning LLMs and inference optimization </how-to/rocm-for-ai/fine-tuning/fine-tuning-and-inference>`.
 
-- To learn how to fine-tune LLMs, see
+- For a list of other ready-made Docker images for AI with ROCm, see
-  :doc:`Fine-tuning LLMs <../../fine-tuning/index>`.
+  `AMD Infinity Hub <https://www.amd.com/en/developer/resources/infinity-hub.html#f-amd_hub_category=AI%20%26%20ML%20Models>`_.
 
 Previous versions
 =================
 
 See :doc:`previous-versions/vllm-history` to find documentation for previous releases
-of the ``ROCm/vllm`` Docker image.
+of the ``ROCm/vllm`` Docker image.

docs/how-to/rocm-for-ai/inference/index.rst

@@ -14,14 +14,14 @@ Throughout the following topics, this section provides a comprehensive guide to
 The AI Developer Hub contains `AMD ROCm tutorials <https://rocm.docs.amd.com/projects/ai-developer-hub/en/latest/>`_ for
 training, fine-tuning, and inference. It leverages popular machine learning frameworks on AMD GPUs.
 
-- :doc:`Installing ROCm and machine learning frameworks <install>`
+- :doc:`Installing ROCm and machine learning frameworks <../install>`
 
 - :doc:`Running models from Hugging Face <hugging-face-models>`
 
 - :doc:`LLM inference frameworks <llm-inference-frameworks>`
 
-- :doc:`vLLM inference performance testing <vllm-benchmark>`
+- :doc:`vLLM inference performance testing <benchmark-docker/vllm>`
 
-- :doc:`PyTorch inference performance testing <pytorch-inference-benchmark>`
+- :doc:`PyTorch inference performance testing <benchmark-docker/pytorch-inference>`
 
 - :doc:`Deploying your model <deploy-your-model>`

docs/how-to/rocm-for-ai/inference/llm-inference-frameworks.rst

@@ -141,7 +141,7 @@ Installing vLLM
 
    ROCm provides a prebuilt optimized Docker image for validating the performance of LLM inference with vLLM
    on the MI300X accelerator. The Docker image includes ROCm, vLLM, and PyTorch.
-   For more information, see :doc:`vllm-benchmark`.
+   For more information, see :doc:`/how-to/rocm-for-ai/inference/benchmark-docker/vllm`.
 
 .. _fine-tuning-llms-tgi:
 

docs/how-to/rocm-for-ai/install.rst

@@ -28,6 +28,8 @@ ROCm supports multiple :doc:`installation methods <rocm-install-on-linux:install
 
 * :doc:`Using your Linux distribution's package manager <rocm-install-on-linux:install/install-methods/package-manager-index>`
 
+* :doc:`Using the AMDGPU installer <rocm-install-on-linux:install/install-methods/amdgpu-installer-index>`
+
 * :ref:`Multi-version installation <rocm-install-on-linux:installation-types>`
 
 .. grid:: 1

docs/how-to/rocm-for-ai/training/benchmark-docker/megatron-lm.rst

@@ -15,57 +15,51 @@ purpose-built to support models like Llama, DeepSeek, and Mixtral,
 enabling developers to train next-generation AI models more
 efficiently.
 
-AMD provides a ready-to-use Docker image for MI300X series accelerators containing
+AMD provides ready-to-use Docker images for MI300X series accelerators containing
 essential components, including PyTorch, ROCm libraries, and Megatron-LM
 utilities. It contains the following software components to accelerate training
 workloads:
 
-+--------------------------+--------------------------------+
-| Software component       | Version                        |
-+==========================+================================+
-| ROCm                     | 6.3.4                          |
-+--------------------------+--------------------------------+
-| PyTorch                  | 2.8.0a0+gite2f9759             |
-+--------------------------+--------------------------------+
-| Python                   | 3.12 or 3.10                   |
-+--------------------------+--------------------------------+
-| Transformer Engine       | 1.13.0+bb061ade                |
-+--------------------------+--------------------------------+
-| Flash Attention          | 3.0.0                          |
-+--------------------------+--------------------------------+
-| hipBLASLt                | 0.13.0-4f18bf6                 |
-+--------------------------+--------------------------------+
-| Triton                   | 3.3.0                          |
-+--------------------------+--------------------------------+
-| RCCL                     | 2.22.3                         |
-+--------------------------+--------------------------------+
-
-Megatron-LM provides the following key features to train large language models efficiently:
-
-- Transformer Engine (TE)
-
-- APEX
-
-- GEMM tuning
-
-- Torch.compile
-
-- 3D parallelism: TP + SP + CP
-
-- Distributed optimizer
-
-- Flash Attention (FA) 3
-
-- Fused kernels
-
-- Pre-training
-
-.. _amd-megatron-lm-model-support:
-
-The following models are pre-optimized for performance on AMD Instinct MI300X series accelerators.
-
 .. datatemplate:yaml:: /data/how-to/rocm-for-ai/training/megatron-lm-benchmark-models.yaml
 
+   {% set dockers = data.dockers %}
+   {% if dockers|length > 1 %}
+   .. tab-set::
+
+      {% for docker in data.dockers %}
+      .. tab-item:: ``{{ docker.pull_tag }}``
+         :sync: {{ docker.pull_tag }}
+
+         .. list-table::
+            :header-rows: 1
+
+            * - Software component
+              - Version
+
+            {% for component_name, component_version in docker.components.items() %}
+            * - {{ component_name }}
+              - {{ component_version }}
+
+            {% endfor %}
+      {% endfor %}
+   {% elif dockers|length == 1 %}
+   .. list-table::
+      :header-rows: 1
+
+      * - Software component
+        - Version
+
+      {% for component_name, component_version in docker.components %}
+      * - {{ component_name }}
+        - {{ component_version }}
+
+      {% endfor %}
+   {% endif %}
+
+   .. _amd-megatron-lm-model-support:
+
+   The following models are pre-optimized for performance on AMD Instinct MI300X series accelerators.
+
    Supported models
    ================
 
@@ -73,8 +67,7 @@ The following models are pre-optimized for performance on AMD Instinct MI300X se
    Some instructions, commands, and training recommendations in this documentation might
    vary by model -- select one to get started.
 
-   {% set model_groups = data["megatron-lm_benchmark"].model_groups %}
+   {% set model_groups = data.model_groups %}
-
    .. raw:: html
 
          <div id="vllm-benchmark-ud-params-picker" class="container-fluid">
@@ -82,7 +75,7 @@ The following models are pre-optimized for performance on AMD Instinct MI300X se
              <div class="col-2 me-2 model-param-head">Model</div>
              <div class="row col-10">
       {% for model_group in model_groups %}
-               <div class="col-4 model-param" data-param-k="model-group" data-param-v="{{ model_group.tag }}" tabindex="0">{{ model_group.group }}</div>
+               <div class="col-3 model-param" data-param-k="model-group" data-param-v="{{ model_group.tag }}" tabindex="0">{{ model_group.group }}</div>
       {% endfor %}
              </div>
            </div>
@@ -115,14 +108,14 @@ Performance measurements
 ========================
 
 To evaluate performance, the
-`Performance results with AMD ROCm software <https://www.amd.com/en/developer/resources/rocm-hub/dev-ai/performance-results.html#tabs-a8deaeb413-item-21cea50186-tab>`_
+`Performance results with AMD ROCm software <https://www.amd.com/en/developer/resources/rocm-hub/dev-ai/performance-results.html#tabs-a8deaeb413-item-21cea50186-tab>`__
 page provides reference throughput and latency measurements for training
 popular AI models.
 
 .. important::
 
    The performance data presented in
-   `Performance results with AMD ROCm software <https://www.amd.com/en/developer/resources/rocm-hub/dev-ai/performance-results.html>`_
+   `Performance results with AMD ROCm software <https://www.amd.com/en/developer/resources/rocm-hub/dev-ai/performance-results.html>`__
    only reflects the latest version of this training benchmarking environment.
    The listed measurements should not be interpreted as the peak performance achievable by AMD Instinct MI325X and MI300X accelerators or ROCm software.
 
@@ -155,27 +148,77 @@ image.
 Download the Docker image
 -------------------------
 
-1. Use the following command to pull the Docker image from Docker Hub.
+.. datatemplate:yaml:: /data/how-to/rocm-for-ai/training/megatron-lm-benchmark-models.yaml
 
-   .. tab-set:: 
+   {% set dockers = data.dockers %}
+   1. Use the following command to pull the Docker image from Docker Hub.
 
-      .. tab-item:: Ubuntu 24.04 + Python 3.12
+      {% if dockers|length > 1 %}
+      .. tab-set:: 
 
-         .. code-block:: shell
+         {% for docker in data.dockers %}
+         .. tab-item:: {{ docker.doc_name }}
+            :sync: {{ docker.pull_tag }}
 
-            docker pull rocm/megatron-lm:v25.5_py312
+            .. code-block:: shell
 
-      .. tab-item:: Ubuntu 22.04 + Python 3.10
+               docker pull {{ docker.pull_tag }}
 
-         .. code-block:: shell
+         {% endfor %}
+      {% elif dockers|length == 1 %}
+      {% set docker = dockers[0] %}
+      .. code-block:: shell
 
-            docker pull rocm/megatron-lm:v25.5_py310
+         docker pull {{ docker.pull_tag }}
 
-2. Launch the Docker container.
+      {% endif %}
+   2. Launch the Docker container.
 
-   .. code-block:: shell
+      {% if dockers|length > 1 %}
+      .. tab-set::
 
-      docker run -it --device /dev/dri --device /dev/kfd --device /dev/infiniband --network host --ipc host --group-add video --cap-add SYS_PTRACE --security-opt seccomp=unconfined --privileged -v $HOME:$HOME -v  $HOME/.ssh:/root/.ssh --shm-size 64G --name megatron_training_env rocm/megatron-lm:v25.5
+         {% for docker in data.dockers %}
+         .. tab-item:: {{ docker.doc_name }}
+            :sync: {{ docker.pull_tag }}
+
+            .. code-block:: shell
+
+               docker run -it \
+                   --device /dev/dri \
+                   --device /dev/kfd \
+                   --device /dev/infiniband \
+                   --network host --ipc host \
+                   --group-add video \
+                   --cap-add SYS_PTRACE \
+                   --security-opt seccomp=unconfined \
+                   --privileged \
+                   -v $HOME:$HOME \
+                   -v $HOME/.ssh:/root/.ssh \
+                   --shm-size 128G \
+                   --name megatron_training_env \
+                   {{ docker.pull_tag }}
+
+         {% endfor %}
+      {% elif dockers|length == 1 %}
+      {% set docker = dockers[0] %}
+      .. code-block:: shell
+
+         docker run -it \
+             --device /dev/dri \
+             --device /dev/kfd \
+             --device /dev/infiniband \
+             --network host --ipc host \
+             --group-add video \
+             --cap-add SYS_PTRACE \
+             --security-opt seccomp=unconfined \
+             --privileged \
+             -v $HOME:$HOME \
+             -v $HOME/.ssh:/root/.ssh \
+             --shm-size 128G \
+             --name megatron_training_env \
+             {{ docker.pull_tag }}
+
+      {% endif %}
 
 3. Use these commands if you exit the ``megatron_training_env`` container and need to return to it.
 
@@ -333,6 +376,22 @@ If the tokenizer is not found, it'll be downloaded if publicly available.
 
       TOKENIZER_MODEL=tokenizer/tokenizer.model
 
+.. container:: model-doc pyt_megatron_lm_train_qwen2.5-7b
+
+   The training script uses the ``HuggingFaceTokenizer``. Set ``TOKENIZER_MODEL`` to the appropriate Hugging Face model path.
+
+   .. code-block:: shell
+
+      TOKENIZER_MODEL="Qwen/Qwen2.5-7B"
+
+.. container:: model-doc pyt_megatron_lm_train_qwen2.5-72b
+
+   The training script uses the ``HuggingFaceTokenizer``. Set ``TOKENIZER_MODEL`` to the appropriate Hugging Face model path.
+
+   .. code-block:: shell
+
+      TOKENIZER_MODEL="Qwen/Qwen2.5-72B"
+
 Dataset options
 ---------------
 
@@ -358,7 +417,7 @@ You can use either mock data or real data for training.
 Download the dataset
 ^^^^^^^^^^^^^^^^^^^^
 
-.. container:: model-doc pyt_megatron_lm_train_llama-3.3-70b pyt_megatron_lm_train_llama-3.1-8b pyt_megatron_lm_train_llama-3.1-70b pyt_megatron_lm_train_llama-2-7b pyt_megatron_lm_train_llama-2-70b
+.. container:: model-doc pyt_megatron_lm_train_llama-3.3-70b pyt_megatron_lm_train_llama-3.1-8b pyt_megatron_lm_train_llama-3.1-70b pyt_megatron_lm_train_llama-2-7b pyt_megatron_lm_train_llama-2-70b pyt_megatron_lm_train_llama-3.1-70b-proxy
 
    For Llama models, use the `prepare_dataset.sh
    <https://github.com/ROCm/Megatron-LM/tree/rocm_dev/examples/llama>`_ script
@@ -397,8 +456,8 @@ Download the dataset
       wget https://atp-modelzoo-wlcb-pai.oss-cn-wulanchabu.aliyuncs.com/release/models/pai-megatron-patch/deepseek-datasets/SlimPajama.json
       wget https://atp-modelzoo-wlcb-pai.oss-cn-wulanchabu.aliyuncs.com/release/models/pai-megatron-patch/deepseek-datasets/alpaca_zh-train.json
       wget https://atp-modelzoo-wlcb-pai.oss-cn-wulanchabu.aliyuncs.com/release/models/pai-megatron-patch/deepseek-datasets/alpaca_zh-valid.json
-      wget https://atp-modelzoo-wlcb-pai.oss-cn-wulanchabu.aliyuncs.com/release/models/pai-megatron-patch/deepseek-datasets/mmap_deepseekv2_datasets_text_document.bin
+      cd ..
-      wget https://atp-modelzoo-wlcb-pai.oss-cn-wulanchabu.aliyuncs.com/release/models/pai-megatron-patch/deepseek-datasets/mmap_deepseekv2_datasets_text_document.idx
+      bash tools/run_make_pretraining_dataset_megatron.sh deepseek-datasets/SlimPajama.json DeepSeekV3Tokenizer text deepseek-datasets deepseek-ai/DeepSeek-V3
 
    To train on this data, update the ``DATA_DIR`` variable to point to the location of your dataset.
 
@@ -422,8 +481,8 @@ Download the dataset
       wget https://atp-modelzoo-wlcb-pai.oss-cn-wulanchabu.aliyuncs.com/release/models/pai-megatron-patch/deepseek-datasets/SlimPajama.json
       wget https://atp-modelzoo-wlcb-pai.oss-cn-wulanchabu.aliyuncs.com/release/models/pai-megatron-patch/deepseek-datasets/alpaca_zh-train.json
       wget https://atp-modelzoo-wlcb-pai.oss-cn-wulanchabu.aliyuncs.com/release/models/pai-megatron-patch/deepseek-datasets/alpaca_zh-valid.json
-      wget https://atp-modelzoo-wlcb-pai.oss-cn-wulanchabu.aliyuncs.com/release/models/pai-megatron-patch/deepseek-datasets/mmap_deepseekv2_datasets_text_document.bin
+      cd ..
-      wget https://atp-modelzoo-wlcb-pai.oss-cn-wulanchabu.aliyuncs.com/release/models/pai-megatron-patch/deepseek-datasets/mmap_deepseekv2_datasets_text_document.idx
+      bash tools/run_make_pretraining_dataset_megatron.sh deepseek-datasets/SlimPajama.json DeepSeekV3Tokenizer text deepseek-datasets deepseek-ai/DeepSeek-V3
 
    To train on this data, update the ``DATA_DIR`` variable to point to the location of your dataset.
 
@@ -433,8 +492,6 @@ Download the dataset
 
       DATA_DIR="<path-to>/deepseek-datasets"  # Change to where your dataset is stored
 
-      Ensure that the files are accessible inside the Docker container.
-
 .. container:: model-doc pyt_megatron_lm_train_mixtral-8x7b pyt_megatron_lm_train_mixtral-8x22b-proxy
 
    If you don't already have the dataset, download the Mixtral dataset using the following
@@ -457,6 +514,27 @@ Download the dataset
 
    Ensure that the files are accessible inside the Docker container.
 
+.. container:: model-doc pyt_megatron_lm_train_qwen2.5-7b pyt_megatron_lm_train_qwen2.5-72b
+
+   If you don't already have the dataset, download the Mixtral dataset using the following
+   commands:
+
+   .. code-block:: shell
+
+      mkdir -p temp/qwen-datasets
+      wget https://atp-modelzoo-wlcb-pai.oss-cn-wulanchabu.aliyuncs.com/release/models/pai-megatron-patch/qwen-datasets/wudao_qwenbpe_text_document.bin
+      wget https://atp-modelzoo-wlcb-pai.oss-cn-wulanchabu.aliyuncs.com/release/models/pai-megatron-patch/qwen-datasets/wudao_qwenbpe_text_document.idx
+
+   To train on this data, update the ``DATA_DIR`` variable to point to the location of your dataset.
+
+   .. code-block:: bash
+
+      MOCK_DATA=0 # Train on real data
+
+      DATA_DIR="<path-to>/qwen-datasets"  # Change to where your dataset is stored
+
+   Ensure that the files are accessible inside the Docker container.
+
 Multi-node configuration
 ------------------------
 
@@ -497,27 +575,17 @@ also be passed as command line arguments. Refer to the following example configu
      # Specify which RDMA interfaces to use for communication
      export NCCL_IB_HCA=rdma0,rdma1,rdma2,rdma3,rdma4,rdma5,rdma6,rdma7
 
-Getting started
-===============
-
-The prebuilt Megatron-LM with ROCm training environment allows users to quickly validate
-system performance, conduct training benchmarks, and achieve superior
-performance for models like Llama, DeepSeek, and Mixtral. This container should not be
-expected to provide generalized performance across all training workloads. You
-can expect the container to perform in the model configurations described in
-the following section, but other configurations are not validated by AMD.
-
 .. _amd-megatron-lm-run-training:
 
 Run training
-------------
+============
 
 Use the following example commands to set up the environment, configure
 :ref:`key options <amd-megatron-lm-benchmark-test-vars>`, and run training on
 MI300X series accelerators with the AMD Megatron-LM environment.
 
 Single node training
-^^^^^^^^^^^^^^^^^^^^
+--------------------
 
 .. container:: model-doc pyt_megatron_lm_train_llama-3.3-70b
 
@@ -526,7 +594,20 @@ Single node training
 
    .. code-block:: shell
 
-      TEE_OUTPUT=1 RECOMPUTE=1 SEQ_LENGTH=8192 MBS=2 BS=16 TE_FP8=0 TP=1 PP=1 FSDP=1 MODEL_SIZE=70 TOTAL_ITERS=50 bash examples/llama/train_llama3.sh 
+      TOKENIZER_MODEL=meta-llama/Llama-3.3-70B-Instruct \
+      CKPT_FORMAT=torch_dist \
+      TEE_OUTPUT=1 \
+      RECOMPUTE=1 \
+      SEQ_LENGTH=8192 \
+      MBS=2 \
+      BS=16 \
+      TE_FP8=0 \
+      TP=1 \
+      PP=1 \
+      FSDP=1 \
+      MODEL_SIZE=70 \
+      TOTAL_ITERS=50 \
+      bash examples/llama/train_llama3.sh 
 
    .. note::
 
@@ -535,8 +616,6 @@ Single node training
       parallelism, MCore's distributed optimizer, gradient accumulation fusion,
       or FP16.
 
-      Currently, FSDP is only compatible with BF16 precision.
-
 .. container:: model-doc pyt_megatron_lm_train_llama-3.1-8b
 
    To run training on a single node for Llama 3.1 8B FP8, navigate to the Megatron-LM folder and use the
@@ -544,13 +623,29 @@ Single node training
 
    .. code-block:: shell
 
-      TEE_OUTPUT=1 MBS=2 BS=128 TP=1 TE_FP8=1 SEQ_LENGTH=8192 MODEL_SIZE=8 TOTAL_ITERS=50 bash examples/llama/train_llama3.sh
+      TEE_OUTPUT=1 \
+      MBS=2 \
+      BS=128 \
+      TP=1 \
+      TE_FP8=1 \
+      SEQ_LENGTH=8192 \
+      MODEL_SIZE=8 \
+      TOTAL_ITERS=50 \
+      bash examples/llama/train_llama3.sh
 
    For Llama 3.1 8B BF16, use the following command:
 
    .. code-block:: shell
 
-      TEE_OUTPUT=1 MBS=2 BS=128 TP=1 TE_FP8=0 SEQ_LENGTH=8192 MODEL_SIZE=8 TOTAL_ITERS=50 bash examples/llama/train_llama3.sh
+      TEE_OUTPUT=1 \
+      MBS=2 \
+      BS=128 \
+      TP=1 \
+      TE_FP8=0 \
+      SEQ_LENGTH=8192 \
+      MODEL_SIZE=8 \
+      TOTAL_ITERS=50 \
+      bash examples/llama/train_llama3.sh
 
 .. container:: model-doc pyt_megatron_lm_train_llama-3.1-70b
 
@@ -559,7 +654,18 @@ Single node training
 
    .. code-block:: shell
 
-      TEE_OUTPUT=1 MBS=3 BS=24 TP=1 TE_FP8=0 FSDP=1 RECOMPUTE=1 SEQ_LENGTH=8192 MODEL_SIZE=70 TOTAL_ITERS=50 bash examples/llama/train_llama3.sh
+      CKPT_FORMAT=torch_dist \
+      TEE_OUTPUT=1 \
+      MBS=3 \
+      BS=24 \
+      TP=1 \
+      TE_FP8=0 \
+      FSDP=1 \
+      RECOMPUTE=1 \
+      SEQ_LENGTH=8192 \
+      MODEL_SIZE=70 \
+      TOTAL_ITERS=50 \
+      bash examples/llama/train_llama3.sh
 
    .. note::
 
@@ -568,7 +674,36 @@ Single node training
       parallelism, MCore's distributed optimizer, gradient accumulation fusion,
       or FP16.
 
-      Currently, FSDP is only compatible with BF16 precision.
+.. container:: model-doc pyt_megatron_lm_train_llama-3.1-70b-proxy
+
+   To run the training on a single node for Llama 3.1 70B with proxy, use the following command.
+
+   .. code-block:: shell
+
+      CKPT_FORMAT=torch_dist \
+      TEE_OUTPUT=1 \
+      RECOMPUTE=1 \
+      MBS=3 \
+      BS=24 \
+      TP=1 \
+      TE_FP8=1 \
+      SEQ_LENGTH=8192 \
+      MODEL_SIZE=70 \
+      FSDP=1 \
+      TOTAL_ITERS=10 \
+      NUM_LAYERS=40 \
+      bash examples/llama/train_llama3.sh
+
+   .. note::
+
+      Use two or more nodes to run the *full* Llama 70B model with FP8 precision.
+
+   .. note::
+
+      It is suggested to use ``TP=1`` when FSDP is enabled for higher
+      throughput. FSDP-v2 is not supported with pipeline parallelism, expert
+      parallelism, MCore's distributed optimizer, gradient accumulation fusion,
+      or FP16.
 
 .. container:: model-doc pyt_megatron_lm_train_llama-2-7b
 
@@ -577,13 +712,29 @@ Single node training
 
    .. code-block:: shell
 
-      TEE_OUTPUT=1 MBS=4 BS=256 TP=1 TE_FP8=1 SEQ_LENGTH=4096 MODEL_SIZE=7 TOTAL_ITERS=50 bash examples/llama/train_llama2.sh
+      TEE_OUTPUT=1 \
+      MBS=4 \
+      BS=256 \
+      TP=1 \
+      TE_FP8=1 \
+      SEQ_LENGTH=4096 \
+      MODEL_SIZE=7 \
+      TOTAL_ITERS=50 \
+      bash examples/llama/train_llama2.sh
 
    For Llama 2 7B BF16, use the following command:
 
    .. code-block:: shell
 
-      TEE_OUTPUT=1 MBS=4 BS=256 TP=1 TE_FP8=0 SEQ_LENGTH=4096 MODEL_SIZE=7 TOTAL_ITERS=50 bash examples/llama/train_llama2.sh
+      TEE_OUTPUT=1 \
+      MBS=4 \
+      BS=256 \
+      TP=1 \
+      TE_FP8=0 \
+      SEQ_LENGTH=4096 \
+      MODEL_SIZE=7 \
+      TOTAL_ITERS=50 \
+      bash examples/llama/train_llama2.sh
 
 .. container:: model-doc pyt_megatron_lm_train_llama-2-70b
 
@@ -592,7 +743,18 @@ Single node training
 
    .. code-block:: shell
 
-      TEE_OUTPUT=1 MBS=7 BS=56 TP=1 TE_FP8=0 FSDP=1 RECOMPUTE=1 SEQ_LENGTH=4096 MODEL_SIZE=70 TOTAL_ITERS=50 bash examples/llama/train_llama2.sh
+      CKPT_FORMAT=torch_dist \
+      TEE_OUTPUT=1 \
+      MBS=7 \
+      BS=56 \
+      TP=1 \
+      TE_FP8=0 \
+      FSDP=1 \
+      RECOMPUTE=1 \
+      SEQ_LENGTH=4096 \
+      MODEL_SIZE=70 \
+      TOTAL_ITERS=50 \
+      bash examples/llama/train_llama2.sh
 
    .. note::
 
@@ -601,8 +763,6 @@ Single node training
       parallelism, MCore's distributed optimizer, gradient accumulation fusion,
       or FP16.
 
-      Currently, FSDP is only compatible with BF16 precision.
-
 .. container:: model-doc pyt_megatron_lm_train_deepseek-v3-proxy
 
    To run training on a single node for DeepSeek-V3 (MoE with expert parallel) with 3-layer proxy, 
@@ -610,7 +770,8 @@ Single node training
 
    .. code-block:: shell
 
-      FORCE_BANLANCE=true \
+      export NVTE_FUSED_ATTN_CK=0
+      FORCE_BALANCE=true \
       RUN_ENV=cluster \
       MODEL_SIZE=671B \
       TRAIN_ITERS=50 \
@@ -632,7 +793,15 @@ Single node training
 
    .. code-block:: shell
 
-      GEMM_TUNING=1 PR=bf16 MBS=4 AC=none SEQ_LEN=4096 PAD_LEN=4096 TRAIN_ITERS=50 bash examples/deepseek_v2/train_deepseekv2.sh
+      export NVTE_FUSED_ATTN_CK=0
+      GEMM_TUNING=1 \
+      PR=bf16 \
+      MBS=4 \
+      AC=none \
+      SEQ_LEN=4096 \
+      PAD_LEN=4096 \
+      TRAIN_ITERS=50 \
+      bash examples/deepseek_v2/train_deepseekv2.sh
 
 .. container:: model-doc pyt_megatron_lm_train_mixtral-8x7b
 
@@ -641,7 +810,24 @@ Single node training
 
    .. code-block:: shell
 
-      RECOMPUTE_NUM_LAYERS=0 TEE_OUTPUT=1 MBS=1 GBS=16 TP_SIZE=1 PP_SIZE=1 AC=none PR=bf16 EP_SIZE=8 ETP_SIZE=1 SEQLEN=4096 FORCE_BALANCE=true MOCK_DATA=1 RUN_ENV=cluster MODEL_SIZE=8x7B TRAIN_ITERS=50 bash examples/mixtral/train_mixtral_moe.sh
+      TOKENIZER_MODEL=<path/to/tokenizer/model>
+      RECOMPUTE_NUM_LAYERS=0 \
+      TEE_OUTPUT=1 \
+      MBS=1 \
+      GBS=16 \
+      TP_SIZE=1 \
+      PP_SIZE=1 \
+      AC=none \
+      PR=bf16 \
+      EP_SIZE=8 \
+      ETP_SIZE=1 \
+      SEQLEN=4096 \
+      FORCE_BALANCE=true \
+      MOCK_DATA=1 \
+      RUN_ENV=cluster \
+      MODEL_SIZE=8x7B \
+      TRAIN_ITERS=50 \
+      bash examples/mixtral/train_mixtral_moe.sh
 
 .. container:: model-doc pyt_megatron_lm_train_mixtral-8x22b-proxy
 
@@ -650,10 +836,85 @@ Single node training
 
    .. code-block:: shell
 
-      RECOMPUTE_NUM_LAYERS=4 TEE_OUTPUT=1 MBS=1 GBS=16 TP_SIZE=1 PP_SIZE=1 AC=full NUM_LAYERS=4 PR=bf16 EP_SIZE=8 ETP_SIZE=1 SEQLEN=8192 FORCE_BALANCE=true MOCK_DATA=1 RUN_ENV=cluster MODEL_SIZE=8x22B TRAIN_ITERS=50 bash examples/mixtral/train_mixtral_moe.sh
+      TOKENIZER_MODEL=<path/to/tokenizer/model>
+      RECOMPUTE_NUM_LAYERS=4 \
+      TEE_OUTPUT=1 \
+      MBS=1 \
+      GBS=16 \
+      TP_SIZE=1 \
+      PP_SIZE=1 \
+      AC=full \
+      NUM_LAYERS=4 \
+      PR=bf16 \
+      EP_SIZE=8 \
+      ETP_SIZE=1 \
+      SEQLEN=8192 \
+      FORCE_BALANCE=true \
+      MOCK_DATA=1 \
+      RUN_ENV=cluster \
+      MODEL_SIZE=8x22B \
+      TRAIN_ITERS=50 \
+      bash examples/mixtral/train_mixtral_moe.sh
 
-Multi-node training
+.. container:: model-doc pyt_megatron_lm_train_qwen2.5-7b
-^^^^^^^^^^^^^^^^^^^
+
+   To run training on a single node for Qwen 2.5 7B BF16, use the following
+   command.
+
+   .. code-block:: shell
+
+      bash examples/qwen/train_qwen2.sh TP=1 \
+          CP=1 \
+          PP=1 \
+          MBS=10 \
+          BS=640 \
+          TE_FP8=0 \
+          MODEL_SIZE=7 \
+          SEQ_LENGTH=2048 \
+          TOTAL_ITERS=50 \
+          MOCK_DATA=1 \
+          TOKENIZER_MODEL=Qwen/Qwen2.5-7B
+
+   For FP8, use the following command.
+
+   .. code-block:: shell
+
+      bash examples/qwen/train_qwen2.sh \
+          TP=1 \
+          CP=1 \
+          PP=1 \
+          MBS=10 \
+          BS=640 \
+          TE_FP8=1 \
+          MODEL_SIZE=7 \
+          SEQ_LENGTH=2048 \
+          TOTAL_ITERS=50 \
+          MOCK_DATA=1 \
+          TOKENIZER_MODEL=Qwen/Qwen2.5-7B
+
+.. container:: model-doc pyt_megatron_lm_train_qwen2.5-72b
+
+   To run the training on a single node for Qwen 2.5 72B BF16, use the following command.
+
+   .. code-block:: shell
+
+      bash examples/qwen/train_qwen2.sh \
+          FSDP=1 \
+          CP=1 \
+          PP=1 \
+          MBS=3 \
+          BS=24 \
+          TE_FP8=0 \
+          MODEL_SIZE=72 \
+          SEQ_LENGTH=2048 \
+          TOTAL_ITERS=50 \
+          MOCK_DATA=1 \
+          TOKENIZER_MODEL=Qwen/Qwen2.5-72B \
+          RECOMPUTE_ACTIVATIONS=full \
+          CKPT_FORMAT=torch_dist
+
+Multi-node training examples
+----------------------------
 
 To run training on multiple nodes, launch the Docker container on each node.
 For example, for Llama 3 using a two node setup (``NODE0`` as the master node),
@@ -663,13 +924,33 @@ use these commands.
 
   .. code-block:: shell
 
-     TEE_OUTPUT=1 MBS=2 BS=256 TP=1 TE_FP8=1 SEQ_LENGTH=8192 MODEL_SIZE=8  MASTER_ADDR=IP_NODE0 NNODES=2 NODE_RANK=0 bash examples/llama/train_llama3.sh
+     TEE_OUTPUT=1 \
+     MBS=2 \
+     BS=256 \
+     TP=1 \
+     TE_FP8=1 \
+     SEQ_LENGTH=8192 \
+     MODEL_SIZE=8  \
+     MASTER_ADDR=IP_NODE0 \
+     NNODES=2 \
+     NODE_RANK=0 \
+     bash examples/llama/train_llama3.sh
 
 * On the worker node ``NODE1``:
 
   .. code-block:: shell
 
-     TEE_OUTPUT=1 MBS=2 BS=256 TP=1 TE_FP8=1 SEQ_LENGTH=8192 MODEL_SIZE=8  MASTER_ADDR=IP_NODE0 NNODES=2 NODE_RANK=1 bash examples/llama/train_llama3.sh
+     TEE_OUTPUT=1 \
+     MBS=2 \
+     BS=256 \
+     TP=1 \
+     TE_FP8=1 \
+     SEQ_LENGTH=8192 \
+     MODEL_SIZE=8  \
+     MASTER_ADDR=IP_NODE0 \
+     NNODES=2 \
+     NODE_RANK=1 \
+     bash examples/llama/train_llama3.sh
 
 Or, for DeepSeek-V3, an example script ``train_deepseek_v3_slurm.sh`` is
 provided in

docs/how-to/rocm-for-ai/training/benchmark-docker/previous-versions/jax-maxtext-history.rst

@@ -12,23 +12,23 @@ previous releases of the ``ROCm/jax-training`` Docker image on `Docker Hub <http
 
 .. list-table::
    :header-rows: 1
-   :stub-columns: 1
 
    * - Image version
-     - ROCm version
+     - Components
-     - JAX version
      - Resources
 
-   * - 25.5
+   * - 25.5 (latest)
-     - 6.3.4
+     - 
-     - 0.4.35
+       * ROCm 6.3.4
+       * JAX 0.4.35
      - 
        * :doc:`Documentation <../jax-maxtext>`
-       * `Docker Hub <https://hub.docker.com/layers/rocm/jax-training/maxtext-v25.5/images/sha256-4e0516358a227cae8f552fb866ec07e2edcf244756f02e7b40212abfbab5217b>`_
+       * `Docker Hub <https://hub.docker.com/layers/rocm/jax-training/maxtext-v25.5/images/sha256-4e0516358a227cae8f552fb866ec07e2edcf244756f02e7b40212abfbab5217b>`__
 
    * - 25.4
-     - 6.3.0
+     - 
-     - 0.4.31
+       * ROCm 6.3.0
+       * JAX 0.4.31
      - 
        * :doc:`Documentation <jax-maxtext-v25.4>`
-       * `Docker Hub <https://hub.docker.com/layers/rocm/jax-training/maxtext-v25.4/images/sha256-fb3eb71cd74298a7b3044b7130cf84113f14d518ff05a2cd625c11ea5f6a7b01>`_
+       * `Docker Hub <https://hub.docker.com/layers/rocm/jax-training/maxtext-v25.4/images/sha256-fb3eb71cd74298a7b3044b7130cf84113f14d518ff05a2cd625c11ea5f6a7b01>`__

docs/how-to/rocm-for-ai/training/benchmark-docker/previous-versions/jax-maxtext-v25.4.rst

@@ -96,14 +96,14 @@ This Docker image is optimized for specific model configurations outlined
 as follows. Performance can vary for other training workloads, as AMD
 doesn’t validate configurations and run conditions outside those described.
 
-.. _amd-maxtext-multi-node-setup:
+.. _amd-maxtext-multi-node-setup-v254:
 
 Multi-node setup
 ----------------
 
 For multi-node environments, ensure you have all the necessary packages for
 your network device, such as, RDMA. If you're not using a multi-node setup
-with RDMA, skip ahead to :ref:`amd-maxtext-download-docker`.
+with RDMA, skip ahead to :ref:`amd-maxtext-download-docker-v254`.
 
 1. Install the following packages to build and install the RDMA driver.
 
@@ -168,7 +168,7 @@ with RDMA, skip ahead to :ref:`amd-maxtext-download-docker`.
 
    e. RDMA interface
 
-      Ensure the :ref:`required packages <amd-maxtext-multi-node-setup>` are installed on all nodes.
+      Ensure the :ref:`required packages <amd-maxtext-multi-node-setup-v254>` are installed on all nodes.
       Then, set the RDMA interfaces to use for communication.
 
       .. code-block:: bash
@@ -178,7 +178,7 @@ with RDMA, skip ahead to :ref:`amd-maxtext-download-docker`.
          # If using Mellanox NIC
          export NCCL_IB_HCA=mlx5_0,mlx5_1,mlx5_2,mlx5_3,mlx5_4,mlx5_5,mlx5_8,mlx5_9
 
-.. _amd-maxtext-download-docker:
+.. _amd-maxtext-download-docker-v254:
 
 Download the Docker image
 -------------------------
@@ -195,7 +195,7 @@ Download the Docker image
 
       docker run -it --device /dev/dri --device /dev/kfd --network host --ipc host --group-add video --cap-add SYS_PTRACE --security-opt seccomp=unconfined --privileged -v $HOME/.ssh:/root/.ssh --shm-size 128G --name maxtext_training rocm/jax-training:maxtext-v25.4
 
-.. _amd-maxtext-get-started:
+.. _amd-maxtext-get-started-v254:
 
 Getting started
 ===============

docs/how-to/rocm-for-ai/training/benchmark-docker/previous-versions/megatron-lm-history.rst

@@ -7,41 +7,53 @@ Megatron-LM training performance testing version history
 This table lists previous versions of the ROCm Megatron-LM training Docker image for
 inference performance testing. For detailed information about available models
 for benchmarking, see the version-specific documentation. You can find tagged
-previous releases of the ``ROCm/megatron-lm`` Docker image on `Docker Hub <https://hub.docker.com/r/rocm/megatron-lm/tags>`_.
+previous releases of the ``ROCm/megatron-lm`` Docker image on `Docker Hub <https://hub.docker.com/r/rocm/megatron-lm/tags>`__.
 
 .. list-table::
    :header-rows: 1
-   :stub-columns: 1
 
    * - Image version
-     - ROCm version
+     - Components
-     - PyTorch version
      - Resources
 
-   * - v25.5
+   * - v25.6 (latest)
-     - 6.3.4
-     - 2.8.0a0+gite2f9759
      - 
-       * `Documentation <../megatron-lm>`_
+       * ROCm 6.4.1
-       * `Docker Hub <https://hub.docker.com/layers/rocm/megatron-lm/v25.5_py312/images/sha256-4506f18ba188d24189c6b1f95130b425f52c528a543bb3f420351824edceadc2>`_
+       * PyTorch 2.8.0a0+git7d205b2
+     - 
+       * :doc:`Documentation <../megatron-lm>`
+       * `Docker Hub (py312) <https://hub.docker.com/layers/rocm/megatron-lm/v25.6_py312/images/sha256-482ff906532285bceabdf2bda629bd32cb6174d2d07f4243a736378001b28df0>`__
+       * `Docker Hub (py310) <https://hub.docker.com/layers/rocm/megatron-lm/v25.6_py310/images/sha256-9627bd9378684fe26cb1a10c7dd817868f553b33402e49b058355b0f095568d6>`__
+
+   * - v25.5
+     - 
+       * ROCm 6.3.4
+       * PyTorch 2.8.0a0+gite2f9759
+     - 
+       * :doc:`Documentation <megatron-lm-v25.5>`
+       * `Docker Hub (py312) <https://hub.docker.com/layers/rocm/megatron-lm/v25.5_py312/images/sha256-4506f18ba188d24189c6b1f95130b425f52c528a543bb3f420351824edceadc2>`__
+       * `Docker Hub (py310) <https://hub.docker.com/layers/rocm/megatron-lm/v25.5_py310/images/sha256-743fbf1ceff7a44c4452f938d783a7abf143737d1c15b2b95f6f8a62e0fd048b>`__
 
    * - v25.4
-     - 6.3.0
+     - 
-     - 2.7.0a0+git637433 
+       * ROCm 6.3.0
+       * PyTorch 2.7.0a0+git637433 
      - 
        * :doc:`Documentation <megatron-lm-v25.4>`
-       * `Docker Hub <https://hub.docker.com/layers/rocm/megatron-lm/v25.4/images/sha256-941aa5387918ea91c376c13083aa1e6c9cab40bb1875abbbb73bbb65d8736b3f>`_
+       * `Docker Hub <https://hub.docker.com/layers/rocm/megatron-lm/v25.4/images/sha256-941aa5387918ea91c376c13083aa1e6c9cab40bb1875abbbb73bbb65d8736b3f>`__
 
    * - v25.3
-     - 6.3.0
+     - 
-     - 2.7.0a0+git637433 
+       * ROCm 6.3.0
+       * PyTorch 2.7.0a0+git637433 
      - 
        * :doc:`Documentation <megatron-lm-v25.3>`
-       * `Docker Hub <https://hub.docker.com/layers/rocm/megatron-lm/v25.3/images/sha256-1e6ed9bdc3f4ca397300d5a9907e084ab5e8ad1519815ee1f868faf2af1e04e2>`_
+       * `Docker Hub <https://hub.docker.com/layers/rocm/megatron-lm/v25.3/images/sha256-1e6ed9bdc3f4ca397300d5a9907e084ab5e8ad1519815ee1f868faf2af1e04e2>`__
 
    * - v24.12-dev
-     - 6.1.0
+     - 
-     - 2.4.0
+       * ROCm 6.1.0
+       * PyTorch 2.4.0
      - 
        * :doc:`Documentation <megatron-lm-v24.12-dev>`
-       * `Docker Hub <https://hub.docker.com/layers/rocm/megatron-lm/24.12-dev/images/sha256-5818c50334ce3d69deeeb8f589d83ec29003817da34158ebc9e2d112b929bf2e>`_
+       * `Docker Hub <https://hub.docker.com/layers/rocm/megatron-lm/24.12-dev/images/sha256-5818c50334ce3d69deeeb8f589d83ec29003817da34158ebc9e2d112b929bf2e>`__

docs/how-to/rocm-for-ai/training/benchmark-docker/previous-versions/megatron-lm-v24.12-dev.rst

@@ -102,7 +102,8 @@ the output is ``1``, run the following command to disable NUMA auto-balancing.
 
    sudo sh -c 'echo 0 > /proc/sys/kernel/numa_balancing'
 
-See :ref:`mi300x-disable-numa` for more information.
+See :ref:`System validation and optimization <rocm-for-ai-system-optimization>`
+for more information.
 
 Hardware verification with ROCm
 -------------------------------
@@ -118,7 +119,7 @@ Run the command:
 
    rocm-smi --setperfdeterminism 1900
 
-See :ref:`mi300x-hardware-verification-with-rocm` for more information.
+See `Hardware verification with ROCm <https://instinct.docs.amd.com/projects/amdgpu-docs/en/latest/system-optimization/mi300x.html#hardware-verification-with-rocm>`_ for more information.
 
 RCCL Bandwidth Test
 -------------------
@@ -171,7 +172,7 @@ Run on 8 GPUs (``-g 8``), scanning from 8 bytes to 10 GB:
 
    ./build/all_reduce_perf -b 8 -e 10G -f 2 -g 8
 
-.. image:: ../../data/how-to/rocm-for-ai/rccl-tests-8-gpu.png
+.. image:: /data/how-to/rocm-for-ai/rccl-tests-8-gpu.png
    :width: 800
 
 Using one MPI process per GPU and ``-g 1`` for performance-oriented runs on both single-node and multi-node is
@@ -181,7 +182,7 @@ recommended. So, a run on 8 GPUs looks something like:
 
    mpirun -np 8 --bind-to numa ./build/all_reduce_perf -b 8 -e 10G -f 2 -g 1
 
-.. image:: ../../data/how-to/rocm-for-ai/rccl-tests-1-mpi-process-per-gpu.png
+.. image:: /data/how-to/rocm-for-ai/rccl-tests-1-mpi-process-per-gpu.png
    :width: 800
 
 Running with one MPI process per GPU ensures a one-to-one mapping for CPUs and GPUs, which can be beneficial
@@ -202,10 +203,10 @@ Use the following script to run the RCCL test for four MI300X GPU nodes. Modify
    -x NCCL_DEBUG=version \
    $HOME/rccl-tests/build/all_reduce_perf -b 8 -e 8g -f 2 -g 1
 
-.. image:: ../../data/how-to/rocm-for-ai/rccl-tests-4-mi300x-gpu-nodes.png
+.. image:: /data/how-to/rocm-for-ai/rccl-tests-4-mi300x-gpu-nodes.png
    :width: 800
 
-.. _mi300x-amd-megatron-lm-training:
+.. _mi300x-amd-megatron-lm-training-v2412:
 
 Start training on MI300X accelerators
 =====================================
@@ -217,7 +218,7 @@ Use the following instructions to set up the environment, configure the script t
 reproduce the benchmark results on the MI300X accelerators with the AMD Megatron-LM Docker
 image.
 
-.. _amd-megatron-lm-requirements:
+.. _amd-megatron-lm-requirements-v2412:
 
 Download the Docker image and required packages
 -----------------------------------------------
@@ -271,10 +272,10 @@ end-of-document token, remove sentence splitting, and use the tokenizer type.
 In this case, the automatically generated output files are named ``my-gpt2_text_document.bin`` and
 ``my-gpt2_text_document.idx``.
 
-.. image:: ../../data/how-to/rocm-for-ai/prep-training-datasets-my-gpt2-text-document.png
+.. image:: /data/how-to/rocm-for-ai/prep-training-datasets-my-gpt2-text-document.png
    :width: 800
 
-.. _amd-megatron-lm-environment-setup:
+.. _amd-megatron-lm-environment-setup-v2412:
 
 Environment setup
 -----------------
@@ -374,19 +375,19 @@ Run benchmark tests
 
         NODE_RANK="${NODE_RANK:-0}"
 
-* Use this command to run a performance benchmark test of any of the Llama 2 models that this Docker image supports (see :ref:`variables <amd-megatron-lm-benchmark-test-vars>`).
+* Use this command to run a performance benchmark test of any of the Llama 2 models that this Docker image supports (see :ref:`variables <amd-megatron-lm-benchmark-test-vars-v2412>`).
 
   .. code-block:: shell
 
      {variables} bash examples/llama/train_llama2.sh
 
-* Use this command to run a performance benchmark test of any of the Llama 3 and Llama 3.1 models that this Docker image supports (see :ref:`variables <amd-megatron-lm-benchmark-test-vars>`).
+* Use this command to run a performance benchmark test of any of the Llama 3 and Llama 3.1 models that this Docker image supports (see :ref:`variables <amd-megatron-lm-benchmark-test-vars-v2412>`).
 
   .. code-block:: shell
 
      {variables} bash examples/llama/train_llama3.sh
 
-.. _amd-megatron-lm-benchmark-test-vars:
+.. _amd-megatron-lm-benchmark-test-vars-v2412:
 
 The benchmark tests support the same set of variables:
 
@@ -465,11 +466,11 @@ Benchmarking examples
          TEE_OUTPUT=1 MBS=5 BS=120 TP=8 TE_FP8=0 NO_TORCH_COMPILE=1
          SEQ_LENGTH=4096 bash examples/llama/train_llama2.sh
 
-      You can find the training logs at the location defined in ``$TRAIN_LOG`` in the :ref:`configuration script <amd-megatron-lm-environment-setup>`.
+      You can find the training logs at the location defined in ``$TRAIN_LOG`` in the :ref:`configuration script <amd-megatron-lm-environment-setup-v2412>`.
 
       See the sample output:
 
-      .. image:: ../../data/how-to/rocm-for-ai/llama2-7b-training-log-sample.png
+      .. image:: /data/how-to/rocm-for-ai/llama2-7b-training-log-sample.png
          :width: 800
 
    .. tab-item:: Multi node training
@@ -494,18 +495,18 @@ Benchmarking examples
          TEE_OUTPUT=1 MBS=4 BS=64 TP=8 TE_FP8=0 NO_TORCH_COMPILE=1
          SEQ_LENGTH=4096 bash examples/llama/train_llama2.sh
 
-      You can find the training logs at the location defined in ``$TRAIN_LOG`` in the :ref:`configuration script <amd-megatron-lm-environment-setup>`.
+      You can find the training logs at the location defined in ``$TRAIN_LOG`` in the :ref:`configuration script <amd-megatron-lm-environment-setup-v2412>`.
 
       Sample output for 2-node training:
 
       Master node:
 
-      .. image:: ../../data/how-to/rocm-for-ai/2-node-training-master.png
+      .. image:: /data/how-to/rocm-for-ai/2-node-training-master.png
          :width: 800
 
       Worker node:
 
-      .. image:: ../../data/how-to/rocm-for-ai/2-node-training-worker.png
+      .. image:: /data/how-to/rocm-for-ai/2-node-training-worker.png
          :width: 800
 
 Previous versions

docs/how-to/rocm-for-ai/training/benchmark-docker/previous-versions/megatron-lm-v25.3.rst

@@ -111,9 +111,10 @@ the output is ``1``, run the following command to disable NUMA auto-balancing.
 
    sudo sh -c 'echo 0 > /proc/sys/kernel/numa_balancing'
 
-See :ref:`mi300x-disable-numa` for more information.
+See :ref:`System validation and optimization <rocm-for-ai-system-optimization>`
+for more information.
 
-.. _mi300x-amd-megatron-lm-training:
+.. _mi300x-amd-megatron-lm-training-v253:
 
 Environment setup
 =================
@@ -125,7 +126,7 @@ Use the following instructions to set up the environment, configure the script t
 reproduce the benchmark results on the MI300X accelerators with the AMD Megatron-LM Docker
 image.
 
-.. _amd-megatron-lm-requirements:
+.. _amd-megatron-lm-requirements-v253:
  
 Download the Docker image
 -------------------------
@@ -151,7 +152,7 @@ Download the Docker image
 
 The Docker container includes a pre-installed, verified version of Megatron-LM from the `release branch <https://github.com/ROCm/Megatron-LM/tree/megatron_release_v25.3>`_.
 
-.. _amd-megatron-lm-environment-setup:
+.. _amd-megatron-lm-environment-setup-v253:
 
 Configuration scripts
 ---------------------
@@ -395,7 +396,7 @@ accelerators with the AMD Megatron-LM Docker image.
 Key options
 -----------
 
-.. _amd-megatron-lm-benchmark-test-vars:
+.. _amd-megatron-lm-benchmark-test-vars-v253:
 
 The benchmark tests support the following sets of variables:
 
@@ -485,11 +486,11 @@ Benchmarking examples
                TEE_OUTPUT=1 MBS=5 BS=120 TP=8 TE_FP8=0 NO_TORCH_COMPILE=1
                SEQ_LENGTH=4096 bash examples/llama/train_llama2.sh
 
-            You can find the training logs at the location defined in ``$TRAIN_LOG`` in the :ref:`configuration script <amd-megatron-lm-environment-setup>`.
+            You can find the training logs at the location defined in ``$TRAIN_LOG`` in the :ref:`configuration script <amd-megatron-lm-environment-setup-v253>`.
 
             See the sample output:
 
-            .. image:: ../../../../data/how-to/rocm-for-ai/llama2-7b-training-log-sample.png
+            .. image:: /data/how-to/rocm-for-ai/llama2-7b-training-log-sample.png
                :width: 800
 
          .. tab-item:: Multi-node training
@@ -514,18 +515,18 @@ Benchmarking examples
                TEE_OUTPUT=1 MBS=4 BS=64 TP=8 TE_FP8=0 NO_TORCH_COMPILE=1
                SEQ_LENGTH=4096 bash examples/llama/train_llama2.sh
 
-            You can find the training logs at the location defined in ``$TRAIN_LOG`` in the :ref:`configuration script <amd-megatron-lm-environment-setup>`.
+            You can find the training logs at the location defined in ``$TRAIN_LOG`` in the :ref:`configuration script <amd-megatron-lm-environment-setup-v253>`.
 
             Sample output for 2-node training:
 
             Master node:
 
-            .. image:: ../../../../data/how-to/rocm-for-ai/2-node-training-master.png
+            .. image:: /data/how-to/rocm-for-ai/2-node-training-master.png
                :width: 800
 
             Worker node:
 
-            .. image:: ../../../../data/how-to/rocm-for-ai/2-node-training-worker.png
+            .. image:: /data/how-to/rocm-for-ai/2-node-training-worker.png
                :width: 800
 
 Previous versions

docs/how-to/rocm-for-ai/training/benchmark-docker/previous-versions/megatron-lm-v25.4.rst

@@ -90,21 +90,21 @@ The following models are pre-optimized for performance on AMD Instinct MI300X se
    Some models, such as Llama, require an external license agreement through
    a third party (for example, Meta).
 
-.. _amd-megatron-lm-performance-measurements:
+.. _amd-megatron-lm-performance-measurements-v254:
 
 Performance measurements
 ========================
 
 To evaluate performance, the
-`Performance results with AMD ROCm software <https://www.amd.com/en/developer/resources/rocm-hub/dev-ai/performance-results.html#tabs-a8deaeb413-item-21cea50186-tab>`_
+`Performance results with AMD ROCm software <https://www.amd.com/en/developer/resources/rocm-hub/dev-ai/performance-results.html#tabs-a8deaeb413-item-21cea50186-tab>`__
 page provides reference throughput and latency measurements for training
 popular AI models.
 
 .. important::
 
    The performance data presented in
-   `Performance results with AMD ROCm software <https://www.amd.com/en/developer/resources/rocm-hub/dev-ai/performance-results.html>`_
+   `Performance results with AMD ROCm software <https://www.amd.com/en/developer/resources/rocm-hub/dev-ai/performance-results.html>`__
-   only reflects the :doc:`latest version of this training benchmarking environment <../megatron-lm>`_.
+   only reflects the :doc:`latest version of this training benchmarking environment <../megatron-lm>`.
    The listed measurements should not be interpreted as the peak performance achievable by AMD Instinct MI325X and MI300X accelerators or ROCm software.
 
 System validation
@@ -115,7 +115,7 @@ auto-balancing, skip this step. Otherwise, complete the :ref:`system validation
 and optimization steps <train-a-model-system-validation>` to set up your system
 before starting training.
 
-.. _mi300x-amd-megatron-lm-training:
+.. _mi300x-amd-megatron-lm-training-v254:
 
 Environment setup
 =================
@@ -127,7 +127,7 @@ Use the following instructions to set up the environment, configure the script t
 reproduce the benchmark results on MI300X series accelerators with the AMD Megatron-LM Docker
 image.
 
-.. _amd-megatron-lm-requirements:
+.. _amd-megatron-lm-requirements-v254:
  
 Download the Docker image
 -------------------------
@@ -154,7 +154,7 @@ Download the Docker image
 The Docker container includes a pre-installed, verified version of the ROCm Megatron-LM development branch `<https://github.com/ROCm/Megatron-LM/tree/rocm_dev>`__
 (commit `fd6f01 <https://github.com/ROCm/Megatron-LM/tree/fd6f0d11d7f9480ace32f22eb7e4dab5314fa350>`_).
 
-.. _amd-megatron-lm-environment-setup:
+.. _amd-megatron-lm-environment-setup-v254:
 
 Configuration scripts
 ---------------------
@@ -468,7 +468,7 @@ accelerators with the AMD Megatron-LM Docker image.
 Key options
 -----------
 
-.. _amd-megatron-lm-benchmark-test-vars:
+.. _amd-megatron-lm-benchmark-test-vars-v254:
 
 The benchmark tests support the following sets of variables:
 
@@ -568,11 +568,11 @@ Benchmarking examples
                TEE_OUTPUT=1 MBS=5 BS=120 TP=8 TE_FP8=0 NO_TORCH_COMPILE=1
                SEQ_LENGTH=4096 bash examples/llama/train_llama2.sh
 
-            You can find the training logs at the location defined in ``$TRAIN_LOG`` in the :ref:`configuration script <amd-megatron-lm-environment-setup>`.
+            You can find the training logs at the location defined in ``$TRAIN_LOG`` in the :ref:`configuration script <amd-megatron-lm-environment-setup-v254>`.
 
             See the sample output:
 
-            .. image:: ../../../../data/how-to/rocm-for-ai/llama2-7b-training-log-sample.png
+            .. image:: /data/how-to/rocm-for-ai/llama2-7b-training-log-sample.png
                :width: 800
 
          .. tab-item:: Multi-node training
@@ -597,18 +597,18 @@ Benchmarking examples
                TEE_OUTPUT=1 MBS=4 BS=64 TP=8 TE_FP8=0 NO_TORCH_COMPILE=1
                SEQ_LENGTH=4096 bash examples/llama/train_llama2.sh
 
-            You can find the training logs at the location defined in ``$TRAIN_LOG`` in the :ref:`configuration script <amd-megatron-lm-environment-setup>`.
+            You can find the training logs at the location defined in ``$TRAIN_LOG`` in the :ref:`configuration script <amd-megatron-lm-environment-setup-v254>`.
 
             Sample output for 2-node training:
 
             Master node:
 
-            .. image:: ../../../../data/how-to/rocm-for-ai/2-node-training-master.png
+            .. image:: /data/how-to/rocm-for-ai/2-node-training-master.png
                :width: 800
 
             Worker node:
 
-            .. image:: ../../../../data/how-to/rocm-for-ai/2-node-training-worker.png
+            .. image:: /data/how-to/rocm-for-ai/2-node-training-worker.png
                :width: 800
 
 Previous versions

docs/how-to/rocm-for-ai/training/benchmark-docker/previous-versions/megatron-lm-v25.5.rst

@@ -0,0 +1,775 @@
+:orphan:
+
+.. meta::
+   :description: How to train a model using Megatron-LM for ROCm.
+   :keywords: ROCm, AI, LLM, train, Megatron-LM, megatron, Llama, tutorial, docker, torch
+
+******************************************
+Training a model with Megatron-LM for ROCm
+******************************************
+
+.. caution::
+
+   This documentation does not reflect the latest version of ROCm Megatron-LM
+   training performance documentation. See :doc:`../megatron-lm` for the latest version.
+
+The `Megatron-LM framework for ROCm <https://github.com/ROCm/Megatron-LM>`_ is
+a specialized fork of the robust Megatron-LM, designed to enable efficient
+training of large-scale language models on AMD GPUs. By leveraging AMD
+Instinct™ MI300X series accelerators, Megatron-LM delivers enhanced
+scalability, performance, and resource utilization for AI workloads. It is
+purpose-built to support models like Llama, DeepSeek, and Mixtral,
+enabling developers to train next-generation AI models more
+efficiently.
+
+AMD provides a ready-to-use Docker image for MI300X series accelerators containing
+essential components, including PyTorch, ROCm libraries, and Megatron-LM
+utilities. It contains the following software components to accelerate training
+workloads:
+
++--------------------------+--------------------------------+
+| Software component       | Version                        |
++==========================+================================+
+| ROCm                     | 6.3.4                          |
++--------------------------+--------------------------------+
+| PyTorch                  | 2.8.0a0+gite2f9759             |
++--------------------------+--------------------------------+
+| Python                   | 3.12 or 3.10                   |
++--------------------------+--------------------------------+
+| Transformer Engine       | 1.13.0+bb061ade                |
++--------------------------+--------------------------------+
+| Flash Attention          | 3.0.0                          |
++--------------------------+--------------------------------+
+| hipBLASLt                | 0.13.0-4f18bf6                 |
++--------------------------+--------------------------------+
+| Triton                   | 3.3.0                          |
++--------------------------+--------------------------------+
+| RCCL                     | 2.22.3                         |
++--------------------------+--------------------------------+
+
+Megatron-LM provides the following key features to train large language models efficiently:
+
+- Transformer Engine (TE)
+
+- APEX
+
+- GEMM tuning
+
+- Torch.compile
+
+- 3D parallelism: TP + SP + CP
+
+- Distributed optimizer
+
+- Flash Attention (FA) 3
+
+- Fused kernels
+
+- Pre-training
+
+.. _amd-megatron-lm-model-support-v255:
+
+The following models are pre-optimized for performance on AMD Instinct MI300X series accelerators.
+
+.. datatemplate:yaml:: /data/how-to/rocm-for-ai/training/previous-versions/megatron-lm-v25.5-benchmark-models.yaml
+
+   Supported models
+   ================
+
+   The following models are supported for training performance benchmarking with Megatron-LM and ROCm.
+   Some instructions, commands, and training recommendations in this documentation might
+   vary by model -- select one to get started.
+
+   {% set model_groups = data["megatron-lm_benchmark"].model_groups %}
+
+   .. raw:: html
+
+         <div id="vllm-benchmark-ud-params-picker" class="container-fluid">
+           <div class="row">
+             <div class="col-2 me-2 model-param-head">Model</div>
+             <div class="row col-10">
+      {% for model_group in model_groups %}
+               <div class="col-4 model-param" data-param-k="model-group" data-param-v="{{ model_group.tag }}" tabindex="0">{{ model_group.group }}</div>
+      {% endfor %}
+             </div>
+           </div>
+
+           <div class="row mt-1">
+             <div class="col-2 me-2 model-param-head">Model variant</div>
+             <div class="row col-10">
+      {% for model_group in model_groups %}
+         {% set models = model_group.models %}
+         {% for model in models %}
+            {% if models|length % 3 == 0 %}
+               <div class="col-4 model-param" data-param-k="model" data-param-v="{{ model.mad_tag }}" data-param-group="{{ model_group.tag }}" tabindex="0">{{ model.model }}</div>
+            {% else %}
+               <div class="col-6 model-param" data-param-k="model" data-param-v="{{ model.mad_tag }}" data-param-group="{{ model_group.tag }}" tabindex="0">{{ model.model }}</div>
+            {% endif %}
+         {% endfor %}
+      {% endfor %}
+             </div>
+           </div>
+         </div>
+
+.. note::
+
+   Some models, such as Llama, require an external license agreement through
+   a third party (for example, Meta).
+
+.. _amd-megatron-lm-performance-measurements-v255:
+
+Performance measurements
+========================
+
+To evaluate performance, the
+`Performance results with AMD ROCm software <https://www.amd.com/en/developer/resources/rocm-hub/dev-ai/performance-results.html#tabs-a8deaeb413-item-21cea50186-tab>`__
+page provides reference throughput and latency measurements for training
+popular AI models.
+
+.. important::
+
+   The performance data presented in
+   `Performance results with AMD ROCm software <https://www.amd.com/en/developer/resources/rocm-hub/dev-ai/performance-results.html>`__
+   only reflects the latest version of this training benchmarking environment.
+   The listed measurements should not be interpreted as the peak performance achievable by AMD Instinct MI325X and MI300X accelerators or ROCm software.
+
+System validation
+=================
+
+Before running AI workloads, it's important to validate that your AMD hardware is configured
+correctly and performing optimally.
+
+If you have already validated your system settings, including aspects like NUMA auto-balancing, you
+can skip this step. Otherwise, complete the procedures in the :ref:`System validation and
+optimization <rocm-for-ai-system-optimization>` guide to properly configure your system settings
+before starting training.
+
+To test for optimal performance, consult the recommended :ref:`System health benchmarks
+<rocm-for-ai-system-health-bench>`. This suite of tests will help you verify and fine-tune your
+system's configuration.
+
+.. _mi300x-amd-megatron-lm-training-v255:
+
+Environment setup
+=================
+
+Use the following instructions to set up the environment, configure the script to train models, and
+reproduce the benchmark results on MI300X series accelerators with the AMD Megatron-LM Docker
+image.
+
+.. _amd-megatron-lm-requirements-v255:
+ 
+Download the Docker image
+-------------------------
+
+1. Use the following command to pull the Docker image from Docker Hub.
+
+   .. tab-set:: 
+
+      .. tab-item:: Ubuntu 24.04 + Python 3.12
+         :sync: py312
+
+         .. code-block:: shell
+
+            docker pull rocm/megatron-lm:v25.5_py312
+
+      .. tab-item:: Ubuntu 22.04 + Python 3.10
+         :sync: py310
+
+         .. code-block:: shell
+
+            docker pull rocm/megatron-lm:v25.5_py310
+
+2. Launch the Docker container.
+
+   .. tab-set::
+
+      .. tab-item:: Ubuntu 24.04 + Python 3.12
+         :sync: py312
+
+         .. code-block:: shell
+
+            docker run -it --device /dev/dri --device /dev/kfd --device /dev/infiniband --network host --ipc host --group-add video --cap-add SYS_PTRACE --security-opt seccomp=unconfined --privileged -v $HOME:$HOME -v  $HOME/.ssh:/root/.ssh --shm-size 128G --name megatron_training_env rocm/megatron-lm:v25.5_py312
+
+
+      .. tab-item:: Ubuntu 22.04 + Python 3.10
+         :sync: py310
+
+         .. code-block:: shell
+
+            docker run -it --device /dev/dri --device /dev/kfd --device /dev/infiniband --network host --ipc host --group-add video --cap-add SYS_PTRACE --security-opt seccomp=unconfined --privileged -v $HOME:$HOME -v  $HOME/.ssh:/root/.ssh --shm-size 128G --name megatron_training_env rocm/megatron-lm:v25.5_py310
+
+3. Use these commands if you exit the ``megatron_training_env`` container and need to return to it.
+
+   .. code-block:: shell
+
+      docker start megatron_training_env
+      docker exec -it megatron_training_env bash
+
+The Docker container includes a pre-installed, verified version of the ROCm
+Megatron-LM development branch
+`<https://github.com/ROCm/Megatron-LM/tree/rocm_dev>`__, including necessary
+training scripts.
+
+.. _amd-megatron-lm-environment-setup-v255:
+
+Configuration
+=============
+
+.. container:: model-doc pyt_megatron_lm_train_llama-3.3-70b pyt_megatron_lm_train_llama-3.1-8b pyt_megatron_lm_train_llama-3.1-70b
+
+   Update the ``train_llama3.sh`` configuration script in the ``examples/llama``
+   directory of
+   `<https://github.com/ROCm/Megatron-LM/tree/rocm_dev/examples/llama>`__ to configure your training run.
+   Options can also be passed as command line arguments as described in :ref:`Run training <amd-megatron-lm-run-training-v255>`.
+
+.. container:: model-doc pyt_megatron_lm_train_llama-2-7b pyt_megatron_lm_train_llama-2-70b
+
+   Update the ``train_llama2.sh`` configuration script in the ``examples/llama``
+   directory of
+   `<https://github.com/ROCm/Megatron-LM/tree/rocm_dev/examples/llama>`__ to configure your training run.
+   Options can also be passed as command line arguments as described in :ref:`Run training <amd-megatron-lm-run-training-v255>`.
+
+.. container:: model-doc pyt_megatron_lm_train_deepseek-v3-proxy
+
+   Update the ``train_deepseekv3.sh`` configuration script in the ``examples/deepseek_v3``
+   directory of
+   `<https://github.com/ROCm/Megatron-LM/tree/rocm_dev/examples/deepseek_v3>`__ to configure your training run.
+   Options can also be passed as command line arguments as described in :ref:`Run training <amd-megatron-lm-run-training-v255>`.
+
+.. container:: model-doc pyt_megatron_lm_train_deepseek-v2-lite-16b
+
+   Update the ``train_deepseekv2.sh`` configuration script in the ``examples/deepseek_v2``
+   directory of
+   `<https://github.com/ROCm/Megatron-LM/tree/rocm_dev/examples/deepseek_v2>`__ to configure your training run.
+   Options can also be passed as command line arguments as described in :ref:`Run training <amd-megatron-lm-run-training-v255>`.
+
+.. container:: model-doc pyt_megatron_lm_train_mixtral-8x7b pyt_megatron_lm_train_mixtral-8x22b-proxy
+
+   Update the ``train_mixtral_moe.sh`` configuration script in the ``examples/mixtral``
+   directory of
+   `<https://github.com/ROCm/Megatron-LM/tree/rocm_dev/examples/mixtral>`__ to configure your training run.
+   Options can also be passed as command line arguments as described in :ref:`Run training <amd-megatron-lm-run-training-v255>`.
+
+.. note::
+
+   See :ref:`Key options <amd-megatron-lm-benchmark-test-vars-v255>` for more information on configuration options.
+
+Network interface
+-----------------
+
+Update the network interface in the script to match your system's network interface. To
+find your network interface, run the following (outside of any Docker container):
+
+.. code-block:: bash
+
+   ip a
+
+Look for an active interface that has an IP address in the same subnet as
+your other nodes. Then, update the following variables in the script, for
+example:
+
+.. code-block:: bash
+
+   export NCCL_SOCKET_IFNAME=ens50f0np0
+
+   export GLOO_SOCKET_IFNAME=ens50f0np0
+
+.. _amd-megatron-lm-tokenizer-v255:
+
+Tokenizer
+---------
+
+You can assign the path of an existing tokenizer to the ``TOKENIZER_MODEL`` as shown in the following examples.
+If the tokenizer is not found, it'll be downloaded if publicly available.
+
+.. container:: model-doc pyt_megatron_lm_train_llama-3.3-70b
+
+   If you do not have Llama 3.3 tokenizer locally, you need to use your
+   personal Hugging Face access token ``HF_TOKEN`` to download the tokenizer.
+   See `Llama-3.3-70B-Instruct
+   <https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct>`_. After you are
+   authorized, use your ``HF_TOKEN`` to download the tokenizer and set the
+   variable ``TOKENIZER_MODEL`` to the tokenizer path.
+
+   .. code-block:: shell
+
+      export HF_TOKEN=<Your personal Hugging Face access token>
+
+   The training script uses the ``HuggingFaceTokenizer``. Set ``TOKENIZER_MODEL`` to the appropriate Hugging Face model path.
+
+   .. code-block:: shell
+
+      TOKENIZER_MODEL="meta-llama/Llama-3.3-70B-Instruct"
+
+.. container:: model-doc pyt_megatron_lm_train_llama-3.1-8b
+
+   The training script uses the ``HuggingFaceTokenizer``. Set ``TOKENIZER_MODEL`` to the appropriate Hugging Face model path.
+
+   .. code-block:: shell
+
+      TOKENIZER_MODEL="meta-llama/Llama-3.1-8B"
+
+.. container:: model-doc pyt_megatron_lm_train_llama-3.1-70b
+
+   The training script uses the ``HuggingFaceTokenizer``. Set ``TOKENIZER_MODEL`` to the appropriate Hugging Face model path.
+
+   .. code-block:: shell
+
+      TOKENIZER_MODEL="meta-llama/Llama-3.1-70B"
+
+.. container:: model-doc pyt_megatron_lm_train_llama-2-7b pyt_megatron_lm_train_llama-2-70b
+
+   The training script uses either the ``Llama2Tokenizer`` or ``HuggingFaceTokenizer`` by default.
+
+.. container:: model-doc pyt_megatron_lm_train_deepseek-v3-proxy
+
+   The training script uses the ``HuggingFaceTokenizer``. Set ``TOKENIZER_MODEL`` to the appropriate Hugging Face model path.
+
+   .. code-block:: shell
+
+      TOKENIZER_MODEL="deepseek-ai/DeepSeek-V3"
+
+.. container:: model-doc pyt_megatron_lm_train_deepseek-v2-lite-16b
+
+   The training script uses the ``HuggingFaceTokenizer``. Set ``TOKENIZER_MODEL`` to the appropriate Hugging Face model path.
+
+   .. code-block:: shell
+
+      TOKENIZER_MODEL="deepseek-ai/DeepSeek-V2-Lite"
+
+.. container:: model-doc pyt_megatron_lm_train_mixtral-8x7b pyt_megatron_lm_train_mixtral-8x22b-proxy
+
+   Download the Mixtral tokenizer.
+
+   .. code-block:: shell
+
+      mkdir tokenizer
+      cd tokenizer
+      export HF_TOKEN=<Your personal Hugging Face access token>
+      wget --header="Authorization: Bearer $HF_TOKEN" -O ./tokenizer.model https://huggingface.co/mistralai/Mixtral-8x7B-v0.1/resolve/main/tokenizer.model
+
+   Use the ``HuggingFaceTokenizer``. Set ``TOKENIZER_MODEL`` to the appropriate Hugging Face model path.
+
+   .. code-block:: shell
+
+      TOKENIZER_MODEL=tokenizer/tokenizer.model
+
+Dataset options
+---------------
+
+You can use either mock data or real data for training.
+
+* Mock data can be useful for testing and validation. Use the ``MOCK_DATA`` variable to toggle between mock and real data. The default
+  value is ``1`` for enabled.
+
+  .. code-block:: bash
+
+     MOCK_DATA=1
+
+* If you're using a real dataset, update the ``DATA_PATH`` variable to point to the location of your dataset.
+
+  .. code-block:: bash
+
+     MOCK_DATA=0
+
+     DATA_PATH="/data/bookcorpus_text_sentence"  # Change to where your dataset is stored
+
+  Ensure that the files are accessible inside the Docker container.
+
+Download the dataset
+^^^^^^^^^^^^^^^^^^^^
+
+.. container:: model-doc pyt_megatron_lm_train_llama-3.3-70b pyt_megatron_lm_train_llama-3.1-8b pyt_megatron_lm_train_llama-3.1-70b pyt_megatron_lm_train_llama-2-7b pyt_megatron_lm_train_llama-2-70b
+
+   For Llama models, use the `prepare_dataset.sh
+   <https://github.com/ROCm/Megatron-LM/tree/rocm_dev/examples/llama>`_ script
+   to prepare your dataset.
+   To download the dataset, set the ``DATASET`` variable to the dataset you'd
+   like to use. Three datasets are supported: ``DATASET=wiki``, ``DATASET=fineweb``, and
+   ``DATASET=bookcorpus``.
+
+   .. code-block:: shell
+
+      DATASET=wiki TOKENIZER_MODEL=NousResearch/Llama-2-7b-chat-hf bash examples/llama/prepare_dataset.sh #for wiki-en dataset
+      DATASET=bookcorpus TOKENIZER_MODEL=NousResearch/Llama-2-7b-chat-hf bash examples/llama/prepare_dataset.sh #for bookcorpus dataset
+
+   ``TOKENIZER_MODEL`` can be any accessible Hugging Face tokenizer.
+   Remember to either pre-download the tokenizer or setup Hugging Face access
+   otherwise when needed -- see the :ref:`Tokenizer <amd-megatron-lm-tokenizer-v255>` section.
+
+   .. note::
+
+      When training set ``DATA_PATH`` to the specific file name prefix pointing to the ``.bin`` or ``.idx``
+      as in the following example:
+
+      .. code-block:: shell
+
+         DATA_PATH="data/bookcorpus_text_sentence" # Change to where your dataset is stored.
+
+.. container:: model-doc pyt_megatron_lm_train_deepseek-v3-proxy
+
+   If you don't already have the dataset, download the DeepSeek dataset using the following
+   commands:
+
+   .. code-block:: shell
+
+      mkdir deepseek-datasets
+      cd deepseek-datasets
+      wget https://atp-modelzoo-wlcb-pai.oss-cn-wulanchabu.aliyuncs.com/release/models/pai-megatron-patch/deepseek-datasets/SlimPajama.json
+      wget https://atp-modelzoo-wlcb-pai.oss-cn-wulanchabu.aliyuncs.com/release/models/pai-megatron-patch/deepseek-datasets/alpaca_zh-train.json
+      wget https://atp-modelzoo-wlcb-pai.oss-cn-wulanchabu.aliyuncs.com/release/models/pai-megatron-patch/deepseek-datasets/alpaca_zh-valid.json
+      wget https://atp-modelzoo-wlcb-pai.oss-cn-wulanchabu.aliyuncs.com/release/models/pai-megatron-patch/deepseek-datasets/mmap_deepseekv2_datasets_text_document.bin
+      wget https://atp-modelzoo-wlcb-pai.oss-cn-wulanchabu.aliyuncs.com/release/models/pai-megatron-patch/deepseek-datasets/mmap_deepseekv2_datasets_text_document.idx
+
+   To train on this data, update the ``DATA_DIR`` variable to point to the location of your dataset.
+
+   .. code-block:: bash
+
+      MOCK_DATA=0 # Train on real data
+
+      DATA_DIR="<path-to>/deepseek-datasets"  # Change to where your dataset is stored
+
+      Ensure that the files are accessible inside the Docker container.
+
+.. container:: model-doc pyt_megatron_lm_train_deepseek-v2-lite-16b
+
+   If you don't already have the dataset, download the DeepSeek dataset using the following
+   commands:
+
+   .. code-block:: shell
+
+      mkdir deepseek-datasets
+      cd deepseek-datasets
+      wget https://atp-modelzoo-wlcb-pai.oss-cn-wulanchabu.aliyuncs.com/release/models/pai-megatron-patch/deepseek-datasets/SlimPajama.json
+      wget https://atp-modelzoo-wlcb-pai.oss-cn-wulanchabu.aliyuncs.com/release/models/pai-megatron-patch/deepseek-datasets/alpaca_zh-train.json
+      wget https://atp-modelzoo-wlcb-pai.oss-cn-wulanchabu.aliyuncs.com/release/models/pai-megatron-patch/deepseek-datasets/alpaca_zh-valid.json
+      wget https://atp-modelzoo-wlcb-pai.oss-cn-wulanchabu.aliyuncs.com/release/models/pai-megatron-patch/deepseek-datasets/mmap_deepseekv2_datasets_text_document.bin
+      wget https://atp-modelzoo-wlcb-pai.oss-cn-wulanchabu.aliyuncs.com/release/models/pai-megatron-patch/deepseek-datasets/mmap_deepseekv2_datasets_text_document.idx
+
+   To train on this data, update the ``DATA_DIR`` variable to point to the location of your dataset.
+
+   .. code-block:: bash
+
+      MOCK_DATA=0 # Train on real data
+
+      DATA_DIR="<path-to>/deepseek-datasets"  # Change to where your dataset is stored
+
+      Ensure that the files are accessible inside the Docker container.
+
+.. container:: model-doc pyt_megatron_lm_train_mixtral-8x7b pyt_megatron_lm_train_mixtral-8x22b-proxy
+
+   If you don't already have the dataset, download the Mixtral dataset using the following
+   commands:
+
+   .. code-block:: shell
+
+      mkdir mixtral-datasets
+      cd mixtral-datasets
+      wget https://atp-modelzoo-wlcb-pai.oss-cn-wulanchabu.aliyuncs.com/release/models/pai-megatron-patch/mistral-datasets/wudao_mistralbpe_content_document.bin
+      wget https://atp-modelzoo-wlcb-pai.oss-cn-wulanchabu.aliyuncs.com/release/models/pai-megatron-patch/mistral-datasets/wudao_mistralbpe_content_document.idx
+
+   To train on this data, update the ``DATA_DIR`` variable to point to the location of your dataset.
+
+   .. code-block:: bash
+
+      MOCK_DATA=0 # Train on real data
+
+      DATA_DIR="<path-to>/mixtral-datasets"  # Change to where your dataset is stored
+
+   Ensure that the files are accessible inside the Docker container.
+
+Multi-node configuration
+------------------------
+
+If you're running multi-node training, update the following environment variables. They can
+also be passed as command line arguments. Refer to the following example configurations.
+
+* Change ``localhost`` to the master node's hostname:
+
+  .. code-block:: shell
+
+     MASTER_ADDR="${MASTER_ADDR:-localhost}"
+
+* Set the number of nodes you want to train on (for instance, ``2``, ``4``, ``8``):
+
+  .. code-block:: shell
+
+     NNODES="${NNODES:-1}"
+
+* Set the rank of each node (0 for master, 1 for the first worker node, and so on):
+
+  .. code-block:: shell
+
+     NODE_RANK="${NODE_RANK:-0}"
+
+* Set ``DATA_CACHE_PATH`` to a common directory accessible by all the nodes (for example, an
+  NFS directory) for multi-node runs:
+
+  .. code-block:: shell
+
+     DATA_CACHE_PATH=/root/cache # Set to a common directory for multi-node runs
+
+* For multi-node runs, make sure the correct network drivers are installed on the nodes. If
+  inside a Docker container, either install the drivers inside the Docker container or pass the network
+  drivers from the host while creating the Docker container.
+
+  .. code-block:: shell
+
+     # Specify which RDMA interfaces to use for communication
+     export NCCL_IB_HCA=rdma0,rdma1,rdma2,rdma3,rdma4,rdma5,rdma6,rdma7
+
+Getting started
+===============
+
+The prebuilt Megatron-LM with ROCm training environment allows users to quickly validate
+system performance, conduct training benchmarks, and achieve superior
+performance for models like Llama, DeepSeek, and Mixtral. This container should not be
+expected to provide generalized performance across all training workloads. You
+can expect the container to perform in the model configurations described in
+the following section, but other configurations are not validated by AMD.
+
+.. _amd-megatron-lm-run-training-v255:
+
+Run training
+------------
+
+Use the following example commands to set up the environment, configure
+:ref:`key options <amd-megatron-lm-benchmark-test-vars-v255>`, and run training on
+MI300X series accelerators with the AMD Megatron-LM environment.
+
+Single node training
+^^^^^^^^^^^^^^^^^^^^
+
+.. container:: model-doc pyt_megatron_lm_train_llama-3.3-70b
+
+   To run the training on a single node for Llama 3.3 70B BF16 with FSDP-v2 enabled, add the ``FSDP=1`` argument.
+   For example, use the following command:
+
+   .. code-block:: shell
+
+      TEE_OUTPUT=1 RECOMPUTE=1 SEQ_LENGTH=8192 MBS=2 BS=16 TE_FP8=0 TP=1 PP=1 FSDP=1 MODEL_SIZE=70 TOTAL_ITERS=50 bash examples/llama/train_llama3.sh 
+
+   .. note::
+
+      It is suggested to use ``TP=1`` when FSDP is enabled for higher
+      throughput. FSDP-v2 is not supported with pipeline parallelism, expert
+      parallelism, MCore's distributed optimizer, gradient accumulation fusion,
+      or FP16.
+
+      Currently, FSDP is only compatible with BF16 precision.
+
+.. container:: model-doc pyt_megatron_lm_train_llama-3.1-8b
+
+   To run training on a single node for Llama 3.1 8B FP8, navigate to the Megatron-LM folder and use the
+   following command.
+
+   .. code-block:: shell
+
+      TEE_OUTPUT=1 MBS=2 BS=128 TP=1 TE_FP8=1 SEQ_LENGTH=8192 MODEL_SIZE=8 TOTAL_ITERS=50 bash examples/llama/train_llama3.sh
+
+   For Llama 3.1 8B BF16, use the following command:
+
+   .. code-block:: shell
+
+      TEE_OUTPUT=1 MBS=2 BS=128 TP=1 TE_FP8=0 SEQ_LENGTH=8192 MODEL_SIZE=8 TOTAL_ITERS=50 bash examples/llama/train_llama3.sh
+
+.. container:: model-doc pyt_megatron_lm_train_llama-3.1-70b
+
+   To run the training on a single node for Llama 3.1 70B BF16 with FSDP-v2 enabled, add the ``FSDP=1`` argument.
+   For example, use the following command:
+
+   .. code-block:: shell
+
+      TEE_OUTPUT=1 MBS=3 BS=24 TP=1 TE_FP8=0 FSDP=1 RECOMPUTE=1 SEQ_LENGTH=8192 MODEL_SIZE=70 TOTAL_ITERS=50 bash examples/llama/train_llama3.sh
+
+   .. note::
+
+      It is suggested to use ``TP=1`` when FSDP is enabled for higher
+      throughput. FSDP-v2 is not supported with pipeline parallelism, expert
+      parallelism, MCore's distributed optimizer, gradient accumulation fusion,
+      or FP16.
+
+      Currently, FSDP is only compatible with BF16 precision.
+
+.. container:: model-doc pyt_megatron_lm_train_llama-2-7b
+
+   To run training on a single node for Llama 2 7B FP8, navigate to the Megatron-LM folder and use the
+   following command.
+
+   .. code-block:: shell
+
+      TEE_OUTPUT=1 MBS=4 BS=256 TP=1 TE_FP8=1 SEQ_LENGTH=4096 MODEL_SIZE=7 TOTAL_ITERS=50 bash examples/llama/train_llama2.sh
+
+   For Llama 2 7B BF16, use the following command:
+
+   .. code-block:: shell
+
+      TEE_OUTPUT=1 MBS=4 BS=256 TP=1 TE_FP8=0 SEQ_LENGTH=4096 MODEL_SIZE=7 TOTAL_ITERS=50 bash examples/llama/train_llama2.sh
+
+.. container:: model-doc pyt_megatron_lm_train_llama-2-70b
+
+   To run the training on a single node for Llama 2 70B BF16 with FSDP-v2 enabled, add the ``FSDP=1`` argument.
+   For example, use the following command:
+
+   .. code-block:: shell
+
+      TEE_OUTPUT=1 MBS=7 BS=56 TP=1 TE_FP8=0 FSDP=1 RECOMPUTE=1 SEQ_LENGTH=4096 MODEL_SIZE=70 TOTAL_ITERS=50 bash examples/llama/train_llama2.sh
+
+   .. note::
+
+      It is suggested to use ``TP=1`` when FSDP is enabled for higher
+      throughput. FSDP-v2 is not supported with pipeline parallelism, expert
+      parallelism, MCore's distributed optimizer, gradient accumulation fusion,
+      or FP16.
+
+      Currently, FSDP is only compatible with BF16 precision.
+
+.. container:: model-doc pyt_megatron_lm_train_deepseek-v3-proxy
+
+   To run training on a single node for DeepSeek-V3 (MoE with expert parallel) with 3-layer proxy, 
+   navigate to the Megatron-LM folder and use the following command.
+
+   .. code-block:: shell
+
+      FORCE_BANLANCE=true \
+      RUN_ENV=cluster \
+      MODEL_SIZE=671B \
+      TRAIN_ITERS=50 \
+      SEQ_LEN=4096 \
+      NUM_LAYERS=3 \
+      MICRO_BATCH_SIZE=1 GLOBAL_BATCH_SIZE=32 \
+      PR=bf16 \
+      TP=1 PP=1 ETP=1 EP=8 \
+      GEMM_TUNING=1 \
+      NVTE_CK_USES_BWD_V3=1 \
+      USE_GROUPED_GEMM=true MOE_USE_LEGACY_GROUPED_GEMM=true \
+      GPT_LAYER_IN_TE=true \
+      bash examples/deepseek_v3/train_deepseekv3.sh
+
+.. container:: model-doc pyt_megatron_lm_train_deepseek-v2-lite-16b
+
+   To run training on a single node for DeepSeek-V2-Lite (MoE with expert parallel),
+   navigate to the Megatron-LM folder and use the following command.
+
+   .. code-block:: shell
+
+      GEMM_TUNING=1 PR=bf16 MBS=4 AC=none SEQ_LEN=4096 PAD_LEN=4096 TRAIN_ITERS=50 bash examples/deepseek_v2/train_deepseekv2.sh
+
+.. container:: model-doc pyt_megatron_lm_train_mixtral-8x7b
+
+   To run training on a single node for Mixtral 8x7B (MoE with expert parallel),
+   navigate to the Megatron-LM folder and use the following command.
+
+   .. code-block:: shell
+
+      RECOMPUTE_NUM_LAYERS=0 TEE_OUTPUT=1 MBS=1 GBS=16 TP_SIZE=1 PP_SIZE=1 AC=none PR=bf16 EP_SIZE=8 ETP_SIZE=1 SEQLEN=4096 FORCE_BALANCE=true MOCK_DATA=1 RUN_ENV=cluster MODEL_SIZE=8x7B TRAIN_ITERS=50 bash examples/mixtral/train_mixtral_moe.sh
+
+.. container:: model-doc pyt_megatron_lm_train_mixtral-8x22b-proxy
+
+   To run training on a single node for Mixtral 8x7B (MoE with expert parallel) with 4-layer proxy,
+   navigate to the Megatron-LM folder and use the following command.
+
+   .. code-block:: shell
+
+      RECOMPUTE_NUM_LAYERS=4 TEE_OUTPUT=1 MBS=1 GBS=16 TP_SIZE=1 PP_SIZE=1 AC=full NUM_LAYERS=4 PR=bf16 EP_SIZE=8 ETP_SIZE=1 SEQLEN=8192 FORCE_BALANCE=true MOCK_DATA=1 RUN_ENV=cluster MODEL_SIZE=8x22B TRAIN_ITERS=50 bash examples/mixtral/train_mixtral_moe.sh
+
+Multi-node training
+^^^^^^^^^^^^^^^^^^^
+
+To run training on multiple nodes, launch the Docker container on each node.
+For example, for Llama 3 using a two node setup (``NODE0`` as the master node),
+use these commands.
+
+* On the master node ``NODE0``:
+
+  .. code-block:: shell
+
+     TEE_OUTPUT=1 MBS=2 BS=256 TP=1 TE_FP8=1 SEQ_LENGTH=8192 MODEL_SIZE=8  MASTER_ADDR=IP_NODE0 NNODES=2 NODE_RANK=0 bash examples/llama/train_llama3.sh
+
+* On the worker node ``NODE1``:
+
+  .. code-block:: shell
+
+     TEE_OUTPUT=1 MBS=2 BS=256 TP=1 TE_FP8=1 SEQ_LENGTH=8192 MODEL_SIZE=8  MASTER_ADDR=IP_NODE0 NNODES=2 NODE_RANK=1 bash examples/llama/train_llama3.sh
+
+Or, for DeepSeek-V3, an example script ``train_deepseek_v3_slurm.sh`` is
+provided in
+`<https://github.com/ROCm/Megatron-LM/tree/rocm_dev/examples/deepseek_v3>`__ to
+enable training at scale under a SLURM environment. For example, to run
+training on 16 nodes, try the following command:
+
+.. code-block:: shell
+
+   sbatch examples/deepseek_v3/train_deepseek_v3_slurm.sh
+
+.. _amd-megatron-lm-benchmark-test-vars-v255:
+
+Key options
+-----------
+
+The benchmark tests support the following sets of variables.
+
+``TEE_OUTPUT``
+  ``1`` to enable training logs or ``0`` to disable.
+
+``TE_FP8``
+  ``0`` for B16 or ``1`` for FP8 -- ``0`` by default.
+
+``GEMM_TUNING``
+  ``1`` to enable GEMM tuning, which boosts performance by using the best GEMM kernels.
+
+``USE_FLASH_ATTN``
+  ``1`` to enable Flash Attention.
+
+``FSDP``
+  ``1`` to enable PyTorch FSDP2. If FSDP is enabled, ``--use-distributed-optimizer``,
+  ``--overlap-param-gather``, and ``--sequence-parallel`` are automatically disabled.
+
+``ENABLE_PROFILING``
+  ``1`` to enable PyTorch profiling for performance analysis.
+
+``transformer-impl``
+  ``transformer_engine`` to use the Transformer Engine (TE) or ``local`` to disable TE.
+
+``MODEL_SIZE``
+  ``8B`` or ``70B`` for Llama 3 and 3.1. ``7B`` or ``70B`` for Llama 2, for example.
+
+``TOTAL_ITERS``
+  The total number of iterations -- ``10`` by default.
+
+``MOCK_DATA``
+  ``1`` to use mock data or ``0`` to use real data you provide.
+
+``MBS``
+  Micro batch size.
+
+``BS``
+  Global batch size.
+
+``TP`` / ``TP_SIZE``
+  Tensor parallel (``1``, ``2``, ``4``, ``8``). ``TP`` is disabled when ``FSDP`` is turned on.
+
+``EP`` / ``EP_SIZE``
+  Expert parallel for MoE models.
+
+``SEQ_LENGTH``
+  Input sequence length.
+
+``PR``
+  Precision for training. ``bf16`` for BF16 (default) or ``fp8`` for FP8 GEMMs.
+
+``AC``
+  Activation checkpointing (``none``, ``sel``, or ``full``) -- ``sel`` by default.
+
+``NUM_LAYERS``
+  Use reduced number of layers as a proxy model.
+
+``RECOMPUTE_NUM_LAYERS``
+  Number of layers used for checkpointing recompute.
+
+Previous versions
+=================
+
+See :doc:`megatron-lm-history` to find documentation for previous releases
+of the ``ROCm/megatron-lm`` Docker image.

docs/how-to/rocm-for-ai/training/benchmark-docker/previous-versions/pytorch-training-history.rst

@@ -11,37 +11,39 @@ previous releases of the ``ROCm/pytorch-training`` Docker image on `Docker Hub <
 
 .. list-table::
    :header-rows: 1
-   :stub-columns: 1
 
    * - Image version
-     - ROCm version
+     - Components
-     - PyTorch version
      - Resources
 
    * - v25.6
-     - 6.3.4
+     - 
-     - 2.8.0a0+git7d205b2
+       * ROCm 6.3.4
+       * PyTorch 2.8.0a0+git7d205b2
      - 
        * :doc:`Documentation <../pytorch-training>`
-       * `Docker Hub <https://hub.docker.com/layers/rocm/pytorch-training/v25.6/images/sha256-a4cea3c493a4a03d199a3e81960ac071d79a4a7a391aa9866add3b30a7842661>`_
+       * `Docker Hub <https://hub.docker.com/layers/rocm/pytorch-training/v25.6/images/sha256-a4cea3c493a4a03d199a3e81960ac071d79a4a7a391aa9866add3b30a7842661>`__
 
    * - v25.5
-     - 6.3.4
+     - 
-     - 2.7.0a0+git637433
+       * ROCm 6.3.4
+       * PyTorch 2.7.0a0+git637433
      - 
        * :doc:`Documentation <pytorch-training-v25.5>`
-       * `Docker Hub <https://hub.docker.com/layers/rocm/pytorch-training/v25.5/images/sha256-d47850a9b25b4a7151f796a8d24d55ea17bba545573f0d50d54d3852f96ecde5>`_
+       * `Docker Hub <https://hub.docker.com/layers/rocm/pytorch-training/v25.5/images/sha256-d47850a9b25b4a7151f796a8d24d55ea17bba545573f0d50d54d3852f96ecde5>`__
 
    * - v25.4
-     - 6.3.0
+     - 
-     - 2.7.0a0+git637433
+       * ROCm 6.3.0
+       * PyTorch 2.7.0a0+git637433
      - 
        * :doc:`Documentation <pytorch-training-v25.4>`
-       * `Docker Hub <https://hub.docker.com/layers/rocm/pytorch-training/v25.4/images/sha256-fa98a9aa69968e654466c06f05aaa12730db79b48b113c1ab4f7a5fe6920a20b>`_
+       * `Docker Hub <https://hub.docker.com/layers/rocm/pytorch-training/v25.4/images/sha256-fa98a9aa69968e654466c06f05aaa12730db79b48b113c1ab4f7a5fe6920a20b>`__
 
    * - v25.3
-     - 6.3.0
+     - 
-     - 2.7.0a0+git637433
+       * ROCm 6.3.0
+       * PyTorch 2.7.0a0+git637433
      - 
        * :doc:`Documentation <pytorch-training-v25.3>`
-       * `Docker Hub <https://hub.docker.com/layers/rocm/pytorch-training/v25.3/images/sha256-0ffdde1b590fd2787b1c7adf5686875b100980b0f314090901387c44253e709b>`_
+       * `Docker Hub <https://hub.docker.com/layers/rocm/pytorch-training/v25.3/images/sha256-0ffdde1b590fd2787b1c7adf5686875b100980b0f314090901387c44253e709b>`__

docs/how-to/rocm-for-ai/training/benchmark-docker/previous-versions/pytorch-training-v25.3.rst

@@ -39,7 +39,7 @@ software components to accelerate training workloads:
 | Triton                   | 3.1                            |
 +--------------------------+--------------------------------+
 
-.. _amd-pytorch-training-model-support:
+.. _amd-pytorch-training-model-support-v253:
 
 Supported models
 ================
@@ -80,7 +80,8 @@ the output is ``1``, run the following command to disable NUMA auto-balancing.
 
    sudo sh -c 'echo 0 > /proc/sys/kernel/numa_balancing'
 
-See :ref:`mi300x-disable-numa` for more information.
+See :ref:`System validation and optimization <rocm-for-ai-system-optimization>`
+for more information.
 
 Environment setup
 =================

docs/how-to/rocm-for-ai/training/benchmark-docker/previous-versions/pytorch-training-v25.4.rst

@@ -39,7 +39,7 @@ software components to accelerate training workloads:
 | Triton                   | 3.1                            |
 +--------------------------+--------------------------------+
 
-.. _amd-pytorch-training-model-support:
+.. _amd-pytorch-training-model-support-v254:
 
 Supported models
 ================
@@ -61,7 +61,7 @@ The following models are pre-optimized for performance on the AMD Instinct MI325
    Some models, such as Llama 3, require an external license agreement through
    a third party (for example, Meta).
 
-.. _amd-pytorch-training-performance-measurements:
+.. _amd-pytorch-training-performance-measurements-v254:
 
 Performance measurements
 ========================

docs/how-to/rocm-for-ai/training/benchmark-docker/previous-versions/pytorch-training-v25.5.rst

@@ -1,3 +1,5 @@
+:orphan:
+
 .. meta::
    :description: How to train a model using PyTorch for ROCm.
    :keywords: ROCm, AI, LLM, train, PyTorch, torch, Llama, flux, tutorial, docker
@@ -38,7 +40,7 @@ software components to accelerate training workloads:
 | Triton                   | 3.2.0                          |
 +--------------------------+--------------------------------+
 
-.. _amd-pytorch-training-model-support:
+.. _amd-pytorch-training-model-support-v255:
 
 Supported models
 ================
@@ -62,7 +64,7 @@ The following models are pre-optimized for performance on the AMD Instinct MI325
    Some models, such as Llama 3, require an external license agreement through
    a third party (for example, Meta).
 
-.. _amd-pytorch-training-performance-measurements:
+.. _amd-pytorch-training-performance-measurements-v255:
 
 Performance measurements
 ========================

docs/how-to/system-optimization/w6000-v620.md

@@ -4,6 +4,7 @@ myst:
     "description": "Learn about system settings and performance tuning for RDNA2-based GPUs."
     "keywords": "RDNA2, workstation, desktop, BIOS, installation, Radeon, pro, v620, w6000"
 ---
+:orphan:
 
 # AMD RDNA2 system optimization
 

docs/how-to/tuning-guides/mi300x/index.rst

@@ -1,3 +1,5 @@
+:orphan:
+
 .. meta::
    :description: How to configure MI300X accelerators to fully leverage their capabilities and achieve optimal performance.
    :keywords: ROCm, AI, machine learning, MI300X, LLM, usage, tutorial, optimization, tuning
@@ -12,8 +14,7 @@ accelerators. They include detailed instructions on system settings and
 application tuning suggestions to help you fully leverage the capabilities of
 these accelerators, thereby achieving optimal performance.
 
-* :doc:`../../rocm-for-ai/inference/vllm-benchmark`
+* :doc:`/how-to/rocm-for-ai/inference-optimization/workload`
-* :doc:`../../rocm-for-ai/inference-optimization/workload`
 * `AMD Instinct MI300X system optimization <https://instinct.docs.amd.com/projects/amdgpu-docs/en/latest/system-optimization/mi300x.html>`_
 
 

docs/sphinx/requirements.txt

@@ -215,9 +215,9 @@ sphinx==8.1.3
     #   sphinx-copybutton
     #   sphinx-design
     #   sphinx-external-toc
+    #   sphinx-last-updated-by-git
     #   sphinx-notfound-page
     #   sphinx-reredirects
-    #   sphinx-sitemap
     #   sphinxcontrib-datatemplates
     #   sphinxcontrib-runcmd
 sphinx-book-theme==1.1.4
@@ -228,11 +228,13 @@ sphinx-design==0.6.1
     # via rocm-docs-core
 sphinx-external-toc==1.0.1
     # via rocm-docs-core
+sphinx-last-updated-by-git==0.3.8
+    # via sphinx-sitemap
 sphinx-notfound-page==1.1.0
     # via rocm-docs-core
 sphinx-reredirects==0.1.6
     # via -r requirements.in
-sphinx-sitemap==2.6.0
+sphinx-sitemap==2.7.2
     # via -r requirements.in
 sphinxcontrib-applehelp==2.0.0
     # via sphinx

docs/what-is-rocm.rst

@@ -98,7 +98,7 @@ System Management
 .. csv-table::
   :header: "Component", "Description"
 
-  ":doc:`AMD SMI <amdsmi:index>`", "C library for Linux that provides a user space interface for applications to monitor and control AMD devices"
+  ":doc:`AMD SMI <amdsmi:index>`", "System management interface to control AMD GPU settings, monitor performance, and retrieve device and process information"
   ":doc:`ROCm Data Center Tool <rdc:index>`", "Simplifies administration and addresses key infrastructure challenges in AMD GPUs in cluster and data-center environments"
   ":doc:`rocminfo <rocminfo:index>`", "Reports system information"
   ":doc:`ROCm SMI <rocm_smi_lib:index>`", "C library for Linux that provides a user space interface for applications to monitor and control GPU applications"