Updating broken link (#5258)

* Update PyTorch training benchmark docker doc to 25.7

* update .wordlist.txt

* update conf.py

* update data sheet

* fix sphinx warnings

.wordlist.txt

@@ -5,6 +5,7 @@ ACEs
 ACS
 AccVGPR
 AccVGPRs
+AITER
 ALU
 AllReduce
 AMD
@@ -115,6 +116,7 @@ Deprecations
 DevCap
 DirectX
 Dockerfile
+Dockerized
 Doxygen
 dropless
 ELMo
@@ -122,6 +124,7 @@ ENDPGM
 EPYC
 ESXi
 EoS
+fas
 FBGEMM
 FFT
 FFTs
@@ -194,6 +197,7 @@ HWE
 HWS
 Haswell
 Higgs
+href
 Hyperparameters
 Huggingface
 ICD
@@ -360,6 +364,7 @@ PowerEdge
 PowerShell
 Pretrained
 Pretraining
+Primus
 Profiler's
 PyPi
 Pytest
@@ -524,6 +529,7 @@ Xilinx
 Xnack
 Xteam
 YAML
+YAMLs
 YML
 YModel
 ZeRO
@@ -584,6 +590,7 @@ completers
 composable
 concretization
 config
+configs
 conformant
 constructible
 convolutional
@@ -794,7 +801,9 @@ preprocessing
 preprocessor
 prequantized
 prerequisites
+pretrain
 pretraining
+primus
 profiler
 profilers
 protobuf
@@ -909,6 +918,7 @@ toolchain
 toolchains
 toolset
 toolsets
+torchtitan
 torchvision
 tqdm
 tracebacks

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

@@ -31,9 +31,9 @@ ROCm Version,6.4.3,6.4.2,6.4.1,6.4.0,6.3.3,6.3.2,6.3.1,6.3.0,6.2.4,6.2.2,6.2.1,6
       :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","2.18.1, 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>`,0.4.35,0.4.35,0.4.35,0.4.35,0.4.31,0.4.31,0.4.31,0.4.31,0.4.26,0.4.26,0.4.26,0.4.26,0.4.26,0.4.26,0.4.26,0.4.26,0.4.26,0.4.26
       :doc:`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,N/A,N/A,0.3.0.post0,N/A,N/A,N/A,N/A,N/A
-      :doc:`Stanford Megatron-LM <../compatibility/ml-compatibility/stanford-megatron-lm-compatibility>`,N/A,N/A,N/A,N/A,85f95ae,85f95ae,85f95ae,85f95ae,N/A,N/A,N/A,N/A,N/A,N/A,N/A,N/A,N/A
+      :doc:`Stanford Megatron-LM <../compatibility/ml-compatibility/stanford-megatron-lm-compatibility>` [#stanford-megatron-lm_compat]_,N/A,N/A,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
       :doc:`DGL <../compatibility/ml-compatibility/dgl-compatibility>` [#dgl_compat]_,N/A,N/A,N/A,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,
-      :doc:`Megablocks <../compatibility/ml-compatibility/megablocks-compatibility>`,N/A,N/A,N/A,N/A,0.7.0,0.7.0,0.7.0,0.7.0,N/A,N/A,N/A,N/A,N/A,N/A,N/A,N/A,N/A
+      :doc:`Megablocks <../compatibility/ml-compatibility/megablocks-compatibility>` [#megablocks_compat]_,N/A,N/A,N/A,N/A,N/A,N/A,N/A,0.7.0,N/A,N/A,N/A,N/A,N/A,N/A,N/A,N/A,N/A
       :doc:`Taichi <../compatibility/ml-compatibility/taichi-compatibility>` [#taichi_compat]_,N/A,N/A,N/A,N/A,N/A,1.8.0b1,N/A,N/A,N/A,N/A,N/A,N/A,N/A,N/A,N/A,N/A,N/A
       `ONNX Runtime <https://onnxruntime.ai/docs/build/eps.html#amd-migraphx>`_,1.2,1.2,1.2,1.2,1.17.3,1.17.3,1.17.3,1.17.3,1.17.3,1.17.3,1.17.3,1.17.3,1.17.3,1.17.3,1.17.3,1.17.3,1.14.1,1.14.1
 ,,,,,,,,,,,,,,,,,,

docs/compatibility/compatibility-matrix.rst

@@ -242,7 +242,9 @@ Expand for full historical view of:
    .. [#mi300_602-past-60] **For ROCm 6.0.2** - MI300A (gfx942) is supported on Ubuntu 22.04.3, RHEL 8.9, and SLES 15 SP5. MI300X (gfx942) is only supported on Ubuntu 22.04.3.
    .. [#mi300_600-past-60] **For ROCm 6.0.0** - MI300A (gfx942) is supported on Ubuntu 22.04.3, RHEL 8.9, and SLES 15 SP5. MI300X (gfx942) is only supported on Ubuntu 22.04.3.
    .. [#verl_compat] verl is only supported on ROCm 6.2.0.
+   .. [#stanford-megatron-lm_compat] Stanford Megatron-LM is only supported on ROCm 6.3.0.
    .. [#dgl_compat] DGL is only supported on ROCm 6.4.0.
+   .. [#megablocks_compat] Megablocks is only supported on ROCm 6.3.0.
    .. [#taichi_compat] Taichi is only supported on ROCm 6.3.2.
    .. [#kfd_support-past-60] As of ROCm 6.4.0, forward and backward compatibility between the AMD Kernel-mode GPU Driver (KMD) and its user space software is provided up to a year apart. For earlier ROCm releases, the compatibility is provided for +/- 2 releases. The tested user space versions on this page were accurate as of the time of initial ROCm release. For the most up-to-date information, see the latest version of this information at `User and kernel-space support matrix <https://rocm.docs.amd.com/projects/install-on-linux/en/latest/reference/user-kernel-space-compat-matrix.html>`_.
    .. [#ROCT-rocr-past-60] Starting from ROCm 6.3.0, the ROCT Thunk Interface is included as part of the ROCr runtime package.

docs/conf.py

@@ -117,11 +117,15 @@ article_pages = [
     {"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/previous-versions/megatron-lm-v25.6", "os": ["linux"]},
+    {"file": "how-to/rocm-for-ai/training/benchmark-docker/previous-versions/megatron-lm-primus-migration-guide", "os": ["linux"]},
+    {"file": "how-to/rocm-for-ai/training/benchmark-docker/primus-megatron", "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/previous-versions/pytorch-training-v25.6", "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"]},
@@ -147,6 +151,8 @@ article_pages = [
     {"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/previous-versions/vllm-0.9.1-20250702", "os": ["linux"]},
+    {"file": "how-to/rocm-for-ai/inference/benchmark-docker/previous-versions/vllm-0.9.1-20250715", "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"]},
 

docs/data/how-to/rocm-for-ai/inference/previous-versions/vllm_0.9.1_20250715-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_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: 0.9.1 (0.9.2.dev364+gb432b7a28.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

@@ -39,7 +39,7 @@ pytorch_inference_benchmark:
         model_repo: Wan-AI/Wan2.1-T2V-14B
         url: https://huggingface.co/Wan-AI/Wan2.1-T2V-14B
         precision: bfloat16
-    - group: Janus-Pro
+    - group: Janus Pro
       tag: janus-pro
       models:
       - model: Janus Pro 7B
@@ -47,3 +47,11 @@ pytorch_inference_benchmark:
         model_repo: deepseek-ai/Janus-Pro-7B
         url: https://huggingface.co/deepseek-ai/Janus-Pro-7B
         precision: bfloat16
+    - group: Hunyuan Video
+      tag: hunyuan
+      models:
+      - model: Hunyuan Video
+        mad_tag: pyt_hy_video
+        model_repo: tencent/HunyuanVideo
+        url: https://huggingface.co/tencent/HunyuanVideo
+        precision: float16

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

@@ -2,11 +2,11 @@ vllm_benchmark:
   unified_docker:
     latest:
       # TODO: update me
-      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
+      pull_tag: rocm/vllm:rocm6.4.1_vllm_0.10.0_20250812
+      docker_hub_url: https://hub.docker.com/layers/rocm/vllm/rocm6.4.1_vllm_0.10.0_20250812/images/sha256-4c277ad39af3a8c9feac9b30bf78d439c74d9b4728e788a419d3f1d0c30cacaa
       rocm_version: 6.4.1
-      vllm_version: 0.9.1 (0.9.2.dev364+gb432b7a28.rocm641)
-      pytorch_version: 2.7.0+gitf717b2a
+      vllm_version: 0.10.0 (0.10.1.dev395+g340ea86df.rocm641)
+      pytorch_version: 2.7.0+gitf717b2a (2.7.0+gitf717b2a)
       hipblaslt_version: 0.15
   model_groups:
     - group: Meta Llama
@@ -27,11 +27,6 @@ vllm_benchmark:
         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
@@ -65,11 +60,6 @@ vllm_benchmark:
         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
@@ -80,72 +70,15 @@ vllm_benchmark:
         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:
@@ -153,11 +86,3 @@ vllm_benchmark:
         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/training/megatron-lm-benchmark-models.yaml

@@ -1,26 +1,15 @@
 dockers:
-  - pull_tag: rocm/megatron-lm:v25.6_py312
-    docker_hub_url: https://hub.docker.com/layers/rocm/megatron-lm/v25.6_py312/images/sha256-482ff906532285bceabdf2bda629bd32cb6174d2d07f4243a736378001b28df0
+  - pull_tag: rocm/megatron-lm:v25.7_py310
+    docker_hub_url: https://hub.docker.com/layers/rocm/megatron-lm/v25.7_py310/images/sha256-6189df849feeeee3ae31bb1e97aef5006d69d2b90c134e97708c19632e20ab5a
     components:
-      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
+      ROCm: 6.4.2
+      Primus: v0.1.0-rc1
+      PyTorch: 2.8.0a0+gitd06a406
       Python: "3.10"
-      Transformer Engine: 2.1.0.dev0+8c4a512
-      hipBLASLt: 393e413
+      Transformer Engine: 2.1.0.dev0+ba586519
+      hipBLASLt: 37ba1d36
       Triton: 3.3.0
-      RCCL: 2.23.4.7a84c5d
-    doc_name: Ubuntu 22.04 + Python 3.10
+      RCCL: 2.22.3
 model_groups:
   - group: Meta Llama
     tag: llama

docs/data/how-to/rocm-for-ai/training/previous-versions/megatron-lm-v25.6-benchmark-models.yaml

@@ -0,0 +1,60 @@
+dockers:
+  - pull_tag: rocm/megatron-lm:v25.6_py312
+    docker_hub_url: https://hub.docker.com/layers/rocm/megatron-lm/v25.6_py312/images/sha256-482ff906532285bceabdf2bda629bd32cb6174d2d07f4243a736378001b28df0
+    components:
+      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
+    tag: deepseek
+    models:
+      - 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
+    tag: mistral
+    models:
+      - model: Mixtral 8x7B
+        mad_tag: pyt_megatron_lm_train_mixtral-8x7b
+      - 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/pytorch-training-v25.6-benchmark-models.yaml

@@ -0,0 +1,120 @@
+unified_docker:
+  latest:
+    pull_tag: rocm/pytorch-training:v25.6
+    docker_hub_url: https://hub.docker.com/r/rocm/pytorch-training/tags
+    rocm_version: 6.4.1
+    pytorch_version: 2.8.0a0+git7d205b2
+    python_version: 3.10.17
+    transformer_engine_version: 1.14.0+2f85f5f2
+    flash_attention_version: 3.0.0.post1
+    hipblaslt_version: 0.15.0-8c6919d
+    triton_version: 3.3.0
+model_groups:
+  - group: Pre-training
+    tag: pre-training
+    models:
+    - model: Llama 3.1 8B
+      mad_tag: pyt_train_llama-3.1-8b
+      model_repo: Llama-3.1-8B
+      url: https://huggingface.co/meta-llama/Llama-3.1-8B
+      precision: BF16
+      training_modes: [pretrain]
+    - model: Llama 3.1 70B
+      mad_tag: pyt_train_llama-3.1-70b
+      model_repo: Llama-3.1-70B
+      url: https://huggingface.co/meta-llama/Llama-3.1-70B-Instruct
+      precision: BF16
+      training_modes: [pretrain]
+    - model: FLUX.1-dev
+      mad_tag: pyt_train_flux
+      model_repo: Flux
+      url: https://huggingface.co/black-forest-labs/FLUX.1-dev
+      precision: BF16
+      training_modes: [pretrain]
+  - group: Fine-tuning
+    tag: fine-tuning
+    models:
+    - model: Llama 4 Scout 17B-16E
+      mad_tag: pyt_train_llama-4-scout-17b-16e
+      model_repo: Llama-4-17B_16E
+      url: https://huggingface.co/meta-llama/Llama-4-Scout-17B-16E
+      precision: BF16
+      training_modes: [finetune_fw, finetune_lora]
+    - model: Llama 3.3 70B
+      mad_tag: pyt_train_llama-3.3-70b
+      model_repo: Llama-3.3-70B
+      url: https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct
+      precision: BF16
+      training_modes: [finetune_fw, finetune_lora, finetune_qlora]
+    - model: Llama 3.2 1B
+      mad_tag: pyt_train_llama-3.2-1b
+      model_repo: Llama-3.2-1B
+      url: https://huggingface.co/meta-llama/Llama-3.2-1B
+      precision: BF16
+      training_modes: [finetune_fw, finetune_lora]
+    - model: Llama 3.2 3B
+      mad_tag: pyt_train_llama-3.2-3b
+      model_repo: Llama-3.2-3B
+      url: https://huggingface.co/meta-llama/Llama-3.2-3B
+      precision: BF16
+      training_modes: [finetune_fw, finetune_lora]
+    - model: Llama 3.2 Vision 11B
+      mad_tag: pyt_train_llama-3.2-vision-11b
+      model_repo: Llama-3.2-Vision-11B
+      url: https://huggingface.co/meta-llama/Llama-3.2-11B-Vision
+      precision: BF16
+      training_modes: [finetune_fw]
+    - model: Llama 3.2 Vision 90B
+      mad_tag: pyt_train_llama-3.2-vision-90b
+      model_repo: Llama-3.2-Vision-90B
+      url: https://huggingface.co/meta-llama/Llama-3.2-90B-Vision
+      precision: BF16
+      training_modes: [finetune_fw]
+    - model: Llama 3.1 8B
+      mad_tag: pyt_train_llama-3.1-8b
+      model_repo: Llama-3.1-8B
+      url: https://huggingface.co/meta-llama/Llama-3.1-8B
+      precision: BF16
+      training_modes: [finetune_fw, finetune_lora]
+    - model: Llama 3.1 70B
+      mad_tag: pyt_train_llama-3.1-70b
+      model_repo: Llama-3.1-70B
+      url: https://huggingface.co/meta-llama/Llama-3.1-70B
+      precision: BF16
+      training_modes: [finetune_fw, finetune_lora, finetune_qlora]
+    - model: Llama 3.1 405B
+      mad_tag: pyt_train_llama-3.1-405b
+      model_repo: Llama-3.1-405B
+      url: https://huggingface.co/meta-llama/Llama-3.1-405B
+      precision: BF16
+      training_modes: [finetune_qlora, HF_finetune_lora]
+    - model: Llama 3 8B
+      mad_tag: pyt_train_llama-3-8b
+      model_repo: Llama-3-8B
+      url: https://huggingface.co/meta-llama/Meta-Llama-3-8B
+      precision: BF16
+      training_modes: [finetune_fw, finetune_lora]
+    - model: Llama 3 70B
+      mad_tag: pyt_train_llama-3-70b
+      model_repo: Llama-3-70B
+      url: https://huggingface.co/meta-llama/Meta-Llama-3-70B
+      precision: BF16
+      training_modes: [finetune_fw, finetune_lora]
+    - model: Llama 2 7B
+      mad_tag: pyt_train_llama-2-7b
+      model_repo: Llama-2-7B
+      url: https://github.com/meta-llama/llama-models/tree/main/models/llama2
+      precision: BF16
+      training_modes: [finetune_fw, finetune_lora, finetune_qlora]
+    - model: Llama 2 13B
+      mad_tag: pyt_train_llama-2-13b
+      model_repo: Llama-2-13B
+      url: https://github.com/meta-llama/llama-models/tree/main/models/llama2
+      precision: BF16
+      training_modes: [finetune_fw, finetune_lora]
+    - model: Llama 2 70B
+      mad_tag: pyt_train_llama-2-70b
+      model_repo: Llama-2-70B
+      url: https://github.com/meta-llama/llama-models/tree/main/models/llama2
+      precision: BF16
+      training_modes: [finetune_lora, finetune_qlora, HF_finetune_lora]

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

@@ -0,0 +1,58 @@
+dockers:
+  - pull_tag: rocm/megatron-lm:v25.7_py310
+    docker_hub_url: https://hub.docker.com/layers/rocm/megatron-lm/v25.7_py310/images/sha256-6189df849feeeee3ae31bb1e97aef5006d69d2b90c134e97708c19632e20ab5a
+    components:
+      ROCm: 6.4.2
+      Primus: v0.1.0-rc1
+      PyTorch: 2.8.0a0+gitd06a406
+      Python: "3.10"
+      Transformer Engine: 2.1.0.dev0+ba586519
+      hipBLASLt: 37ba1d36
+      Triton: 3.3.0
+      RCCL: 2.22.3
+model_groups:
+  - group: Meta Llama
+    tag: llama
+    models:
+      - model: Llama 3.3 70B
+        mad_tag: primus_pyt_megatron_lm_train_llama-3.3-70b
+        config_name: llama3.3_70B-pretrain.yaml
+      - model: Llama 3.1 70B
+        mad_tag: primus_pyt_megatron_lm_train_llama-3.1-70b
+        config_name: llama3.1_70B-pretrain.yaml
+      - model: Llama 3.1 8B
+        mad_tag: primus_pyt_megatron_lm_train_llama-3.1-8b
+        config_name: llama3.1_8B-pretrain.yaml
+      - model: Llama 2 7B
+        mad_tag: primus_pyt_megatron_lm_train_llama-2-7b
+        config_name: llama2_7B-pretrain.yaml
+      - model: Llama 2 70B
+        mad_tag: primus_pyt_megatron_lm_train_llama-2-70b
+        config_name: llama2_70B-pretrain.yaml
+  - group: DeepSeek
+    tag: deepseek
+    models:
+      - model: DeepSeek-V3 (proxy)
+        mad_tag: primus_pyt_megatron_lm_train_deepseek-v3-proxy
+        config_name: deepseek_v3-pretrain.yaml
+      - model: DeepSeek-V2-Lite
+        mad_tag: primus_pyt_megatron_lm_train_deepseek-v2-lite-16b
+        config_name: deepseek_v2_lite-pretrain.yaml
+  - group: Mistral AI
+    tag: mistral
+    models:
+      - model: Mixtral 8x7B
+        mad_tag: primus_pyt_megatron_lm_train_mixtral-8x7b
+        config_name: mixtral_8x7B_v0.1-pretrain.yaml
+      - model: Mixtral 8x22B (proxy)
+        mad_tag: primus_pyt_megatron_lm_train_mixtral-8x22b-proxy
+        config_name: mixtral_8x22B_v0.1-pretrain.yaml
+  - group: Qwen
+    tag: qwen
+    models:
+      - model: Qwen 2.5 7B
+        mad_tag: primus_pyt_megatron_lm_train_qwen2.5-7b
+        config_name: primus_qwen2.5_7B-pretrain.yaml
+      - model: Qwen 2.5 72B
+        mad_tag: primus_pyt_megatron_lm_train_qwen2.5-72b
+        config_name: qwen2.5_72B-pretrain.yaml

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

@@ -1,38 +1,17 @@
-unified_docker:
-  latest:
-    pull_tag: rocm/pytorch-training:v25.6
-    docker_hub_url: https://hub.docker.com/r/rocm/pytorch-training/tags
-    rocm_version: 6.4.1
-    pytorch_version: 2.8.0a0+git7d205b2
-    python_version: 3.10.17
-    transformer_engine_version: 1.14.0+2f85f5f2
-    flash_attention_version: 3.0.0.post1
-    hipblaslt_version: 0.15.0-8c6919d
-    triton_version: 3.3.0
+dockers:
+  - pull_tag: rocm/pytorch-training:v25.7
+    docker_hub_url: https://hub.docker.com/layers/rocm/pytorch-training/v25.7/images/sha256-cc6fd840ab89cb81d926fc29eca6d075aee9875a55a522675a4b9231c9a0a712
+    components:
+      ROCm: 6.4.2
+      PyTorch: 2.8.0a0+gitd06a406
+      Python: 3.10.18
+      Transformer Engine: 2.2.0.dev0+94e53dd8
+      Flash Attention: 3.0.0.post1
+      hipBLASLt: 1.1.0-4b9a52edfc
+      Triton: 3.3.0
 model_groups:
-  - group: Pre-training
-    tag: pre-training
-    models:
-    - model: Llama 3.1 8B
-      mad_tag: pyt_train_llama-3.1-8b
-      model_repo: Llama-3.1-8B
-      url: https://huggingface.co/meta-llama/Llama-3.1-8B
-      precision: BF16
-      training_modes: [pretrain]
-    - model: Llama 3.1 70B
-      mad_tag: pyt_train_llama-3.1-70b
-      model_repo: Llama-3.1-70B
-      url: https://huggingface.co/meta-llama/Llama-3.1-70B-Instruct
-      precision: BF16
-      training_modes: [pretrain]
-    - model: FLUX.1-dev
-      mad_tag: pyt_train_flux
-      model_repo: Flux
-      url: https://huggingface.co/black-forest-labs/FLUX.1-dev
-      precision: BF16
-      training_modes: [pretrain]
-  - group: Fine-tuning
-    tag: fine-tuning
+  - group: Meta Llama
+    tag: llama
     models:
     - model: Llama 4 Scout 17B-16E
       mad_tag: pyt_train_llama-4-scout-17b-16e
@@ -75,19 +54,19 @@ model_groups:
       model_repo: Llama-3.1-8B
       url: https://huggingface.co/meta-llama/Llama-3.1-8B
       precision: BF16
-      training_modes: [finetune_fw, finetune_lora]
+      training_modes: [pretrain, finetune_fw, finetune_lora, HF_pretrain]
     - model: Llama 3.1 70B
       mad_tag: pyt_train_llama-3.1-70b
       model_repo: Llama-3.1-70B
-      url: https://huggingface.co/meta-llama/Llama-3.1-70B
+      url: https://huggingface.co/meta-llama/Llama-3.1-70B-Instruct
       precision: BF16
-      training_modes: [finetune_fw, finetune_lora, finetune_qlora]
+      training_modes: [pretrain, finetune_fw, finetune_lora]
     - model: Llama 3.1 405B
       mad_tag: pyt_train_llama-3.1-405b
       model_repo: Llama-3.1-405B
       url: https://huggingface.co/meta-llama/Llama-3.1-405B
       precision: BF16
-      training_modes: [finetune_qlora, HF_finetune_lora]
+      training_modes: [finetune_qlora]
     - model: Llama 3 8B
       mad_tag: pyt_train_llama-3-8b
       model_repo: Llama-3-8B
@@ -117,4 +96,67 @@ model_groups:
       model_repo: Llama-2-70B
       url: https://github.com/meta-llama/llama-models/tree/main/models/llama2
       precision: BF16
-      training_modes: [finetune_lora, finetune_qlora, HF_finetune_lora]
+      training_modes: [finetune_lora, finetune_qlora]
+  - group: OpenAI
+    tag: openai
+    models:
+    - model: GPT OSS 20B
+      mad_tag: pyt_train_gpt_oss_20b
+      model_repo: GPT-OSS-20B
+      url: https://huggingface.co/openai/gpt-oss-20b
+      precision: BF16
+      training_modes: [HF_finetune_lora]
+    - model: GPT OSS 120B
+      mad_tag: pyt_train_gpt_oss_120b
+      model_repo: GPT-OSS-120B
+      url: https://huggingface.co/openai/gpt-oss-120b
+      precision: BF16
+      training_modes: [HF_finetune_lora]
+  - group: Qwen
+    tag: qwen
+    models:
+    - model: Qwen 3 8B
+      mad_tag: pyt_train_qwen3-8b
+      model_repo: Qwen3-8B
+      url: https://huggingface.co/Qwen/Qwen3-8B
+      precision: BF16
+      training_modes: [finetune_fw, finetune_lora]
+    - model: Qwen 3 32B
+      mad_tag: pyt_train_qwen3-32b
+      model_repo: Qwen3-32
+      url: https://huggingface.co/Qwen/Qwen3-32B
+      precision: BF16
+      training_modes: [finetune_lora]
+    - model: Qwen 2.5 32B
+      mad_tag: pyt_train_qwen2.5-32b
+      model_repo: Qwen2.5-32B
+      url: https://huggingface.co/Qwen/Qwen2.5-32B
+      precision: BF16
+      training_modes: [finetune_lora]
+    - model: Qwen 2.5 72B
+      mad_tag: pyt_train_qwen2.5-72b
+      model_repo: Qwen2.5-72B
+      url: https://huggingface.co/Qwen/Qwen2.5-72B
+      precision: BF16
+      training_modes: [finetune_lora]
+    - model: Qwen 2 1.5B
+      mad_tag: pyt_train_qwen2-1.5b
+      model_repo: Qwen2-1.5B
+      url: https://huggingface.co/Qwen/Qwen2-1.5B
+      precision: BF16
+      training_modes: [finetune_fw, finetune_lora]
+    - model: Qwen 2 7B
+      mad_tag: pyt_train_qwen2-7b
+      model_repo: Qwen2-7B
+      url: https://huggingface.co/Qwen/Qwen2-7B
+      precision: BF16
+      training_modes: [finetune_fw, finetune_lora]
+  - group: Flux
+    tag: flux
+    models:
+    - model: FLUX.1-dev
+      mad_tag: pyt_train_flux
+      model_repo: Flux
+      url: https://huggingface.co/black-forest-labs/FLUX.1-dev
+      precision: BF16
+      training_modes: [pretrain]

docs/how-to/build-rocm.rst

@@ -19,5 +19,6 @@ The general steps to build ROCm are:
 #. Run the build command
 
 Because the ROCm stack is constantly evolving, the most current instructions are stored with the source code in GitHub.  
-For detailed build instructions, see `Getting and Building ROCm from Source <https://github.com/ROCm/ROCm?tab=readme-ov-file#getting-and-building-rocm-from-source>`.
+For detailed build instructions, see `Getting and Building ROCm from Source <https://github.com/ROCm/ROCm?tab=readme-ov-file#getting-and-building-rocm-from-source>`_.
+
 

docs/how-to/deep-learning-rocm.rst

@@ -2,58 +2,124 @@
    :description: How to install deep learning frameworks for ROCm
    :keywords: deep learning, frameworks, ROCm, install, PyTorch, TensorFlow, JAX, MAGMA, DeepSpeed, ML, AI
 
-********************************************
-Installing deep learning frameworks for ROCm
-********************************************
+**********************************
+Deep learning frameworks for ROCm
+**********************************
 
-ROCm provides a comprehensive ecosystem for deep learning development, including
-:ref:`libraries <artificial-intelligence-apis>` for optimized deep learning operations and ROCm-aware versions of popular
-deep learning frameworks and libraries such as PyTorch, TensorFlow, and JAX. ROCm works closely with these
-frameworks to ensure that framework-specific optimizations take advantage of AMD accelerator and GPU architectures.
+Deep learning frameworks provide environments for machine learning, training, fine-tuning, inference, and performance optimization.
 
-The following guides provide information on compatibility and supported
-features for these ROCm-enabled deep learning frameworks.
+ROCm offers a complete ecosystem for developing and running deep learning applications efficiently. It also provides ROCm-compatible versions of popular frameworks and libraries, such as PyTorch, TensorFlow, JAX, and others.
 
-* :doc:`PyTorch compatibility <../compatibility/ml-compatibility/pytorch-compatibility>`
-* :doc:`TensorFlow compatibility <../compatibility/ml-compatibility/tensorflow-compatibility>`
-* :doc:`JAX compatibility <../compatibility/ml-compatibility/jax-compatibility>`
-* :doc:`verl compatibility <../compatibility/ml-compatibility/verl-compatibility>`
-* :doc:`Stanford Megatron-LM compatibility <../compatibility/ml-compatibility/stanford-megatron-lm-compatibility>`
-* :doc:`DGL compatibility <../compatibility/ml-compatibility/dgl-compatibility>`
-* :doc:`Megablocks compatibility <../compatibility/ml-compatibility/megablocks-compatibility>`
-* :doc:`Taichi compatibility <../compatibility/ml-compatibility/taichi-compatibility>`
+The AMD ROCm organization actively contributes to open-source development and collaborates closely with framework organizations. This collaboration ensures that framework-specific optimizations effectively leverage AMD GPUs and accelerators.
 
-This chart steps through typical installation workflows for installing deep learning frameworks for ROCm.
+The table below summarizes information about ROCm-enabled deep learning frameworks. It includes details on ROCm compatibility and third-party tool support, installation steps and options, and links to GitHub resources. For a complete list of supported framework versions on ROCm, see the :doc:`Compatibility matrix <../compatibility/compatibility-matrix>` topic.
 
-.. image:: ../data/how-to/framework_install_2024_07_04.png
-   :alt: Flowchart for installing ROCm-aware machine learning frameworks
-   :align: center
+.. list-table:: 
+    :header-rows: 1
+    :widths: 5 3 6 3
 
-See the installation instructions to get started.
+    * - Framework
+      - Installation
+      - Installation options
+      - GitHub
 
-* :doc:`PyTorch for ROCm <rocm-install-on-linux:install/3rd-party/pytorch-install>`
-* :doc:`TensorFlow for ROCm <rocm-install-on-linux:install/3rd-party/tensorflow-install>`
-* :doc:`JAX for ROCm <rocm-install-on-linux:install/3rd-party/jax-install>`
-* :doc:`verl for ROCm <rocm-install-on-linux:install/3rd-party/verl-install>`
-* :doc:`Stanford Megatron-LM for ROCm <rocm-install-on-linux:install/3rd-party/stanford-megatron-lm-install>`
-* :doc:`DGL for ROCm <rocm-install-on-linux:install/3rd-party/dgl-install>`
-* :doc:`Megablocks for ROCm <rocm-install-on-linux:install/3rd-party/megablocks-install>`
-* :doc:`Taichi for ROCm <rocm-install-on-linux:install/3rd-party/taichi-install>`
+    * - `PyTorch <https://rocm.docs.amd.com/en/latest/compatibility/ml-compatibility/pytorch-compatibility.html>`__
+      - .. raw:: html
 
-.. note::
+          <a href="https://rocm.docs.amd.com/projects/install-on-linux/en/latest/install/3rd-party/pytorch-install.html"><i class="fas fa-link fa-lg"></i></a>
+      - 
+        - `Docker image <https://rocm.docs.amd.com/projects/install-on-linux/en/latest/install/3rd-party/pytorch-install.html#using-a-docker-image-with-pytorch-pre-installed>`__
+        - `Wheels package <https://rocm.docs.amd.com/projects/install-on-linux/en/latest/install/3rd-party/pytorch-install.html#using-a-wheels-package>`__
+        - `ROCm Base Docker image <https://rocm.docs.amd.com/projects/install-on-linux/en/latest/install/3rd-party/pytorch-install.html#using-the-pytorch-rocm-base-docker-image>`__
+        - `Upstream Docker file <https://rocm.docs.amd.com/projects/install-on-linux/en/latest/install/3rd-party/pytorch-install.html#using-the-pytorch-upstream-dockerfile>`__
+      - .. raw:: html
 
-   For guidance on installing ROCm itself, refer to :doc:`ROCm installation for Linux <rocm-install-on-linux:index>`.
+          <a href="https://github.com/ROCm/pytorch"><i class="fab fa-github fa-lg"></i></a>
+
+    * - `TensorFlow <https://rocm.docs.amd.com/en/latest/compatibility/ml-compatibility/tensorflow-compatibility.html>`__
+      - .. raw:: html
+
+          <a href="https://rocm.docs.amd.com/projects/install-on-linux/en/latest/install/3rd-party/tensorflow-install.html"><i class="fas fa-link fa-lg"></i></a>
+      - 
+        - `Docker image <https://rocm.docs.amd.com/projects/install-on-linux/en/latest/install/3rd-party/tensorflow-install.html#using-a-docker-image-with-tensorflow-pre-installed>`__
+        - `Wheels package <https://rocm.docs.amd.com/projects/install-on-linux/en/latest/install/3rd-party/tensorflow-install.html#using-a-wheels-package>`__
+
+      - .. raw:: html
+
+          <a href="https://github.com/ROCm/tensorflow-upstream"><i class="fab fa-github fa-lg"></i></a> 
+
+    * - `JAX <https://rocm.docs.amd.com/en/latest/compatibility/ml-compatibility/jax-compatibility.html>`__
+      - .. raw:: html
+
+          <a href="https://rocm.docs.amd.com/projects/install-on-linux/en/latest/install/3rd-party/jax-install.html"><i class="fas fa-link fa-lg"></i></a>
+      - 
+        - `Docker image <https://rocm.docs.amd.com/projects/install-on-linux/en/latest/install/3rd-party/jax-install.html#using-a-prebuilt-docker-image>`__
+      - .. raw:: html
+
+          <a href="https://github.com/ROCm/jax"><i class="fab fa-github fa-lg"></i></a>
+
+    * - `verl <https://rocm.docs.amd.com/en/latest/compatibility/ml-compatibility/verl-compatibility.html>`__
+      - .. raw:: html
+
+          <a href="https://rocm.docs.amd.com/projects/install-on-linux/en/latest/install/3rd-party/verl-install.html"><i class="fas fa-link fa-lg"></i></a>
+      - 
+        - `Docker image <https://rocm.docs.amd.com/projects/install-on-linux/en/latest/install/3rd-party/verl-install.html#use-a-prebuilt-docker-image-with-verl-pre-installed>`__
+      - .. raw:: html
+
+          <a href="https://github.com/ROCm/verl"><i class="fab fa-github fa-lg"></i></a>
+
+    * - `Stanford Megatron-LM <https://rocm.docs.amd.com/en/latest/compatibility/ml-compatibility/stanford-megatron-lm-compatibility.html>`__
+      - .. raw:: html
+
+          <a href="https://rocm.docs.amd.com/projects/install-on-linux/en/latest/install/3rd-party/stanford-megatron-lm-install.html"><i class="fas fa-link fa-lg"></i></a>
+      - 
+        - `Docker image <https://rocm.docs.amd.com/projects/install-on-linux/en/latest/install/3rd-party/stanford-megatron-lm-install.html#use-a-prebuilt-docker-image-with-stanford-megatron-lm-pre-installed>`__
+      - .. raw:: html
+
+          <a href="https://github.com/ROCm/Stanford-Megatron-LM"><i class="fab fa-github fa-lg"></i></a>
+
+    * - `DGL <https://rocm.docs.amd.com/en/latest/compatibility/ml-compatibility/dgl-compatibility.html>`__
+      - .. raw:: html
+
+          <a href="https://rocm.docs.amd.com/projects/install-on-linux/en/latest/install/3rd-party/dgl-install.html"><i class="fas fa-link fa-lg"></i></a>
+      - 
+        - `Docker image <https://rocm.docs.amd.com/projects/install-on-linux/en/latest/install/3rd-party/dgl-install.html#use-a-prebuilt-docker-image-with-dgl-pre-installed>`__
+      - .. raw:: html
+
+          <a href="https://github.com/ROCm/dgl"><i class="fab fa-github fa-lg"></i></a> 
+
+    * - `Megablocks <https://rocm.docs.amd.com/en/latest/compatibility/ml-compatibility/megablocks-compatibility.html>`__
+      - .. raw:: html
+
+          <a href="https://rocm.docs.amd.com/projects/install-on-linux/en/latest/install/3rd-party/megablocks-install.html"><i class="fas fa-link fa-lg"></i></a>
+      - 
+        - `Docker image <https://rocm.docs.amd.com/projects/install-on-linux/en/latest/install/3rd-party/megablocks-install.html#using-a-prebuilt-docker-image-with-megablocks-pre-installed>`__
+      - .. raw:: html
+
+          <a href="https://github.com/ROCm/megablocks"><i class="fab fa-github fa-lg"></i></a>
+
+    * - `Taichi <https://rocm.docs.amd.com/en/latest/compatibility/ml-compatibility/taichi-compatibility.html>`__
+      - .. raw:: html
+
+          <a href="https://rocm.docs.amd.com/projects/install-on-linux/en/latest/install/3rd-party/taichi-install.html"><i class="fas fa-link fa-lg"></i></a>
+      - 
+        - `Docker image <https://rocm.docs.amd.com/projects/install-on-linux/en/latest/install/3rd-party/taichi-install.html#use-a-prebuilt-docker-image-with-taichi-pre-installed>`__
+        - `Wheels package <https://rocm.docs.amd.com/projects/install-on-linux/en/latest/install/3rd-party/taichi-install.html#use-a-wheels-package>`__
+
+      - .. raw:: html
+
+          <a href="https://github.com/ROCm/taichi"><i class="fab fa-github fa-lg"></i></a>
 
 Learn how to use your ROCm deep learning environment for training, fine-tuning, inference, and performance optimization
 through the following guides.
 
 * :doc:`rocm-for-ai/index`
 
-* :doc:`Training <rocm-for-ai/training/index>`
+* :doc:`Use ROCm for training <rocm-for-ai/training/index>`
 
-* :doc:`Fine-tuning LLMs <rocm-for-ai/fine-tuning/index>`
+* :doc:`Use ROCm for fine-tuning LLMs <rocm-for-ai/fine-tuning/index>`
 
-* :doc:`Inference <rocm-for-ai/inference/index>`
+* :doc:`Use ROCm for AI inference <rocm-for-ai/inference/index>`
 
-* :doc:`Inference optimization <rocm-for-ai/inference-optimization/index>`
+* :doc:`Use ROCm for AI inference optimization <rocm-for-ai/inference-optimization/index>`
 

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

@@ -939,7 +939,7 @@ hipBLASLt benchmarking
 The GEMM library
 `hipBLASLt <https://rocm.docs.amd.com/projects/hipBLASLt/en/latest/index.html>`_
 provides a benchmark tool for its supported operations. Refer to the
-`documentation <https://github.com/ROCm/hipBLASLt/blob/develop/clients/benchmarks/README.md>`_
+`documentation <https://github.com/ROCm/hipBLASLt/blob/develop/clients/bench/README.md>`_
 for details.
 
 * Example 1: Benchmark mix fp8 GEMM

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

@@ -14,7 +14,7 @@ vLLM inference performance testing
    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:
+.. _vllm-benchmark-unified-docker-702:
 
 .. datatemplate:yaml:: /data/how-to/rocm-for-ai/inference/previous-versions/vllm_0.9.1_20250702-benchmark-models.yaml
 
@@ -77,7 +77,7 @@ vLLM inference performance testing
         </div>
       </div>
 
-   .. _vllm-benchmark-vllm:
+   .. _vllm-benchmark-vllm-702:
 
    {% for model_group in model_groups %}
       {% for model in model_group.models %}
@@ -159,7 +159,7 @@ vLLM inference performance testing
    Once the setup is complete, choose between two options to reproduce the
    benchmark results:
 
-   .. _vllm-benchmark-mad:
+   .. _vllm-benchmark-mad-702:
 
    {% for model_group in model_groups %}
       {% for model in model_group.models %}

docs/how-to/rocm-for-ai/inference/benchmark-docker/previous-versions/vllm-0.9.1-20250715.rst

@@ -0,0 +1,450 @@
+: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-715:
+
+.. datatemplate:yaml:: /data/how-to/rocm-for-ai/inference/previous-versions/vllm_0.9.1_20250715-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:
+
+   .. list-table::
+      :header-rows: 1
+
+      * - Software component
+        - Version
+
+      * - `ROCm <https://github.com/ROCm/ROCm>`__
+        - {{ unified_docker.rocm_version }}
+
+      * - `vLLM <https://docs.vllm.ai/en/latest>`__
+        - {{ unified_docker.vllm_version }}
+
+      * - `PyTorch <https://github.com/ROCm/pytorch>`__
+        - {{ unified_docker.pytorch_version }}
+
+      * - `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-715>` for
+MI300X series accelerators.
+
+What's new
+==========
+
+The following is summary of notable changes since the :doc:`previous ROCm/vLLM Docker release <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/previous-versions/vllm_0.9.1_20250715-benchmark_models.yaml
+
+   {% set unified_docker = data.vllm_benchmark.unified_docker.latest %}
+   {% set model_groups = data.vllm_benchmark.model_groups %}
+
+   .. _vllm-benchmark-available-models-715:
+
+   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-715:
+
+   {% 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-715:
+
+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.
+
+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.
+
+.. datatemplate:yaml:: /data/how-to/rocm-for-ai/inference/previous-versions/vllm_0.9.1_20250715-benchmark_models.yaml
+
+   {% set unified_docker = data.vllm_benchmark.unified_docker.latest %}
+   {% set model_groups = data.vllm_benchmark.model_groups %}
+
+   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-715:
+
+   {% 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
+
+            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.
+
+               .. code-block:: shell
+
+                  git clone https://github.com/ROCm/MAD
+                  cd MAD
+                  pip install -r requirements.txt
+
+            2. 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"
+                  madengine run \
+                      --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-715>` 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, include the ``--tunableop on`` argument in your
+               run.
+
+               Enabling TunableOp triggers a two-pass run -- a warm-up followed
+               by the performance-collection run.
+
+            {% endif %}
+
+         .. tab-item:: Standalone benchmarking
+
+            .. rubric:: Download the Docker image and required scripts
+
+            1. Run the vLLM benchmark tool independently by starting the
+               `Docker container <{{ unified_docker.docker_hub_url }}>`_
+               as shown in the following snippet.
+
+               .. code-block:: shell
+
+                  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 $(pwd):/workspace \
+                      --env HUGGINGFACE_HUB_CACHE=/workspace \
+                      --name test \
+                      {{ unified_docker.pull_tag }}
+
+            2. In the Docker container, clone the ROCm MAD repository and navigate to the
+               benchmark scripts directory at ``~/MAD/scripts/vllm``.
+
+               .. code-block:: shell
+
+                  git clone https://github.com/ROCm/MAD
+                  cd MAD/scripts/vllm
+
+            3. To start the benchmark, use the following command with the appropriate options.
+
+               .. dropdown:: Benchmark options
+                  :open:
+
+                  .. list-table::
+                     :header-rows: 1
+                     :align: center
+
+                     * - Name
+                       - Options
+                       - Description
+
+                     * - ``$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
+
+                  The input sequence length, output sequence length, and tensor parallel (TP) are
+                  already configured. You don't need to specify them with this script.
+
+               Command:
+
+               .. code-block::
+
+                  ./vllm_benchmark_report.sh \
+                      -s $test_option \
+                      -m {{model.model_repo}} \
+                      -g $num_gpu \
+                      -d {{model.precision}}
+
+               .. note::
+
+                  For best performance, it's recommend to run with ``VLLM_V1_USE_PREFILL_DECODE_ATTENTION=1``.
+
+                  If you encounter the following error, pass your access-authorized Hugging
+                  Face token to the gated models.
+
+                  .. code-block::
+
+                     OSError: You are trying to access a gated repo.
+
+                     # pass your HF_TOKEN
+                     export HF_TOKEN=$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 :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 %}
+
+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
+===============
+
+- 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 MAD and the ``madengine`` CLI, see the `MAD usage guide <https://github.com/ROCm/MAD?tab=readme-ov-file#usage-guide>`__.
+
+- To learn more about system settings and management practices to configure your system for
+  AMD Instinct 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

@@ -16,14 +16,23 @@ previous releases of the ``ROCm/vllm`` Docker image on `Docker Hub <https://hub.
      - Components
      - Resources
 
-   * - ``rocm/vllm:rocm6.4.1_vllm_0.9.1_20250715``
+   * - ``rocm/vllm:rocm6.4.1_vllm_0.10.0_20250812``
        (latest)
+     - 
+       * ROCm 6.4.1
+       * vLLM 0.10.0
+       * PyTorch 2.7.0
+     - 
+       * :doc:`Documentation <../vllm>`
+       * `Docker Hub <https://hub.docker.com/layers/rocm/vllm/rocm6.4.1_vllm_0.10.0_20250812/images/sha256-4c277ad39af3a8c9feac9b30bf78d439c74d9b4728e788a419d3f1d0c30cacaa>`__
+
+   * - ``rocm/vllm:rocm6.4.1_vllm_0.9.1_20250715``
      - 
        * ROCm 6.4.1
        * vLLM 0.9.1
        * PyTorch 2.7.0
      - 
-       * :doc:`Documentation <../vllm>`
+       * :doc:`Documentation <vllm-0.9.1-20250715>`
        * `Docker Hub <https://hub.docker.com/layers/rocm/vllm/rocm6.4.1_vllm_0.9.1_20250715/images/sha256-4a429705fa95a58f6d20aceab43b1b76fa769d57f32d5d28bd3f4e030e2a78ea>`__
 
    * - ``rocm/vllm:rocm6.4.1_vllm_0.9.1_20250702``

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

@@ -103,7 +103,7 @@ 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 pyt_wan2.1_inference pyt_janus_pro_inference
+   .. container:: model-doc pyt_clip_inference pyt_mochi_video_inference pyt_wan2.1_inference pyt_janus_pro_inference pyt_hy_video
 
       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.
 

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

@@ -7,7 +7,7 @@
 vLLM inference performance testing
 **********************************
 
-.. _vllm-benchmark-unified-docker:
+.. _vllm-benchmark-unified-docker-812:
 
 .. datatemplate:yaml:: /data/how-to/rocm-for-ai/inference/vllm-benchmark-models.yaml
 
@@ -39,7 +39,7 @@ vLLM inference performance testing
         - {{ unified_docker.hipblaslt_version }}
 
 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-812>` for
 MI300X series accelerators.
 
 What's new
@@ -47,17 +47,11 @@ 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.
+* Upgraded to vLLM v0.10.
 
-* 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.
+* FP8 KV cache support via AITER.
 
-* 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.
+* Full graph capture support via AITER.
 
 Supported models
 ================
@@ -67,7 +61,7 @@ Supported models
    {% set unified_docker = data.vllm_benchmark.unified_docker.latest %}
    {% set model_groups = data.vllm_benchmark.model_groups %}
 
-   .. _vllm-benchmark-available-models:
+   .. _vllm-benchmark-available-models-812:
 
    The following models are supported for inference performance benchmarking
    with vLLM and ROCm. Some instructions, commands, and recommendations in this
@@ -102,7 +96,7 @@ Supported models
       </div>
       </div>
 
-   .. _vllm-benchmark-vllm:
+   .. _vllm-benchmark-vllm-812:
 
    {% for model_group in model_groups %}
       {% for model in model_group.models %}
@@ -124,14 +118,14 @@ Supported models
    See :ref:`fine-tuning-llms-vllm` and :ref:`mi300x-vllm-optimization` for
    more information.
 
-.. _vllm-benchmark-performance-measurements:
+.. _vllm-benchmark-performance-measurements-812:
 
 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.
+page provides reference throughput and serving measurements for inferencing popular AI models.
 
 .. important::
 
@@ -176,7 +170,7 @@ system's configuration.
    Once the setup is complete, choose between two options to reproduce the
    benchmark results:
 
-   .. _vllm-benchmark-mad:
+   .. _vllm-benchmark-mad-812:
 
    {% for model_group in model_groups %}
       {% for model in model_group.models %}
@@ -209,12 +203,15 @@ system's configuration.
                       --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}}/``.
+            ``container_ci-{{model.mad_tag}}``. The throughput and serving reports of the
+            model are collected in the following paths: ``{{ model.mad_tag }}_throughput.csv``
+            and ``{{ model.mad_tag }}_serving.csv``.
 
-            Although the :ref:`available models <vllm-benchmark-available-models>` are preconfigured
-            to collect latency and throughput performance data, you can also change the benchmarking
-            parameters. See the standalone benchmarking tab for more information.
+            Although the :ref:`available models
+            <vllm-benchmark-available-models-812>` are preconfigured to collect
+            offline throughput and online serving performance data, you can
+            also change the benchmarking parameters. See the standalone
+            benchmarking tab for more information.
 
             {% if model.tunableop %}
 
@@ -224,14 +221,12 @@ system's configuration.
                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, include the ``--tunableop on`` argument in your
-               run.
+               By default, ``{{model.mad_tag}}`` runs with TunableOp disabled (see
+               `<https://github.com/ROCm/MAD/blob/develop/models.json>`__). To enable it, include
+               the ``--tunableop on`` argument in your run.
 
-               Enabling TunableOp triggers a two-pass run -- a warm-up followed
-               by the performance-collection run.
+               Enabling TunableOp triggers a two-pass run -- a warm-up followed by the
+               performance-collection run.
 
             {% endif %}
 
@@ -269,6 +264,13 @@ system's configuration.
 
             3. To start the benchmark, use the following command with the appropriate options.
 
+               .. code-block::
+
+                  ./run.sh \
+                      --config $CONFIG_CSV \
+                      --model_repo {{ model.model_repo }} \
+                      <overrides>
+
                .. dropdown:: Benchmark options
                   :open:
 
@@ -280,42 +282,40 @@ system's configuration.
                        - Options
                        - Description
 
-                     * - ``$test_option``
-                       - latency
-                       - Measure decoding token latency
+                     * - ``--config``
+                       - ``configs/default.csv``
+                       - Run configs from the CSV for the chosen model repo and benchmark.
 
                      * -
-                       - throughput
-                       - Measure token generation throughput
+                       - ``configs/extended.csv``
+                       - 
 
                      * -
-                       - all
-                       - Measure both throughput and latency
+                       - ``configs/performance.csv``
+                       - 
 
-                     * - ``$num_gpu``
-                       - 1 or 8
-                       - Number of GPUs
+                     * - ``--benchmark``
+                       - ``throughput``
+                       - Measure offline end-to-end throughput.
 
-                     * - ``$datatype``
-                       - ``float16`` or ``float8``
-                       - Data type
+                     * - 
+                       - ``serving``
+                       - Measure online serving performance.
+
+                     * - 
+                       - ``all``
+                       - Measure both throughput and serving.
+
+                     * - `<overrides>`
+                       - See `run.sh <https://github.com/ROCm/MAD/blob/develop/scripts/vllm/run.sh>`__ for more info.
+                       - Additional overrides to the config CSV.
 
                   The input sequence length, output sequence length, and tensor parallel (TP) are
                   already configured. You don't need to specify them with this script.
 
-               Command:
-
-               .. code-block::
-
-                  ./vllm_benchmark_report.sh \
-                      -s $test_option \
-                      -m {{model.model_repo}} \
-                      -g $num_gpu \
-                      -d {{model.precision}}
-
                .. note::
 
-                  For best performance, it's recommend to run with ``VLLM_V1_USE_PREFILL_DECODE_ATTENTION=1``.
+                  For best performance, it's recommended 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.
@@ -331,33 +331,33 @@ system's configuration.
 
             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}}
+                 export MAD_MODEL_NAME={{ model.mad_tag }}
+                 ./run.sh \
+                     --config configs/default.csv \
+                     --model_repo {{model.model_repo}} \
+                     --benchmark throughput
 
-              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``.
+              Find the throughput benchmark report at ``./{{ model.mad_tag }}_throughput.csv``.
+
+            * Serving benchmark
+
+              Use this command to benchmark the serving performance of the {{model.model}} model on eight GPUs with :literal:`{{model.precision}}` precision.
+
+              .. code-block::
+
+                 export MAD_MODEL_NAME={{ model.mad_tag }}
+                 ./run.sh \
+                     --config configs/default.csv \
+                     --model_repo {{model.model_repo}} \
+                     --benchmark serving
+
+              Find the serving benchmark report at ``./{{ model.mad_tag }}_serving.csv``.
 
             .. raw:: html
 
@@ -400,7 +400,7 @@ To reproduce this ROCm/vLLM Docker image release, follow these steps:
    .. code-block:: shell
 
       cd vllm
-      git checkout b432b7a285aa0dcb9677380936ffa74931bb6d6f
+      git checkout 340ea86dfe5955d6f9a9e767d6abab5aacf2c978
 
 3. Build the Docker image. Replace ``vllm-rocm`` with your desired image tag.
 
@@ -408,11 +408,6 @@ To reproduce this ROCm/vLLM Docker image release, follow these steps:
 
       docker build -f docker/Dockerfile.rocm -t vllm-rocm .
 
-Known issues and workarounds
-============================
-
-AITER does not support FP8 KV cache yet.
-
 Further reading
 ===============
 

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

@@ -1,14 +1,14 @@
 .. meta::
-   :description: How to install ROCm and popular machine learning frameworks.
+   :description: How to install ROCm and popular deep learning frameworks.
    :keywords: ROCm, AI, LLM, train, fine-tune, FSDP, DeepSpeed, LLaMA, tutorial
 
 .. _rocm-for-ai-install:
 
-***********************************************
-Installing ROCm and machine learning frameworks
-***********************************************
+********************************************
+Installing ROCm and deep learning frameworks
+********************************************
 
-Before getting started, install ROCm and supported machine learning frameworks.
+Before getting started, install ROCm and supported deep learning frameworks.
 
 .. grid:: 1
 
@@ -22,9 +22,9 @@ If you’re new to ROCm, refer to the :doc:`ROCm quick start install guide for L
 <rocm-install-on-linux:install/quick-start>`.
 
 If you’re using a Radeon GPU for graphics-accelerated applications, refer to the
-`Radeon installation instructions <https://rocm.docs.amd.com/projects/radeon/en/docs-6.1.3/docs/install/native_linux/install-radeon.html>`_.
+`Radeon installation instructions <https://rocm.docs.amd.com/projects/radeon/en/latest/docs/install/native_linux/howto_native_linux.html>`_.
 
-You can install ROCm on :ref:`compatible systems <rocm-install-on-linux:reference/system-requirements>` via your Linux
+You can install ROCm on :doc:`compatible systems <rocm-install-on-linux:reference/system-requirements>` via your Linux
 distribution's package manager. See the following documentation resources to get started:
 
 * :doc:`ROCm installation overview <rocm-install-on-linux:install/install-overview>`
@@ -43,29 +43,16 @@ distribution's package manager. See the following documentation resources to get
       If you encounter any issues during installation, refer to the
       :doc:`Installation troubleshooting <rocm-install-on-linux:reference/install-faq>` guide.
 
-Machine learning frameworks
-===========================
+Deep learning frameworks
+========================
 
-ROCm supports popular machine learning frameworks and libraries including `PyTorch
+ROCm supports deep learning frameworks and libraries including `PyTorch
 <https://pytorch.org/blog/pytorch-for-amd-rocm-platform-now-available-as-python-package>`_, `TensorFlow
-<https://tensorflow.org>`_, `JAX <https://jax.readthedocs.io/en/latest>`_, and `DeepSpeed
-<https://cloudblogs.microsoft.com/opensource/2022/03/21/supporting-efficient-large-model-training-on-amd-instinct-gpus-with-deepspeed/>`_.
+<https://tensorflow.org>`_, `JAX <https://jax.readthedocs.io/en/latest>`_, and more.
 
-Review the framework installation documentation. For ease-of-use, it's recommended to use official ROCm prebuilt Docker
+Review the :doc:`framework installation documentation <../deep-learning-rocm>`. For ease-of-use, it's recommended to use official ROCm prebuilt Docker
 images with the framework pre-installed.
 
-* :doc:`PyTorch for ROCm <rocm-install-on-linux:install/3rd-party/pytorch-install>`
-
-* :doc:`TensorFlow for ROCm <rocm-install-on-linux:install/3rd-party/tensorflow-install>`
-
-* :doc:`JAX for ROCm <rocm-install-on-linux:install/3rd-party/jax-install>`
-
-* :doc:`verl for ROCm <rocm-install-on-linux:install/3rd-party/verl-install>`
-
-* :doc:`Stanford Megatron-LM for ROCm <rocm-install-on-linux:install/3rd-party/jax-install>`
-
-* :doc:`DGL for ROCm <rocm-install-on-linux:install/3rd-party/jax-install>`
-
 Next steps
 ==========
 

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

@@ -1,3 +1,5 @@
+: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
@@ -6,6 +8,14 @@
 Training a model with Megatron-LM for ROCm
 ******************************************
 
+.. caution::
+
+   The ROCm Megatron-LM framework now has limited support with this Docker
+   environment; it now focuses on Primus with Megatron-Core. See :doc:`primus-megatron`.
+
+   To learn how to migrate your existing workloads to Primus with Megatron-Core,
+   see :doc:`previous-versions/megatron-lm-primus-migration-guide`.
+
 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
@@ -20,13 +30,17 @@ essential components, including PyTorch, ROCm libraries, and Megatron-LM
 utilities. It contains the following software components to accelerate training
 workloads:
 
+.. note::
+
+   This Docker environment is based on Python 3.10 and Ubuntu 22.04. For an alternative environment with
+   Python 3.12 and Ubuntu 24.04, see the :doc:`previous ROCm Megatron-LM v25.6 Docker release <previous-versions/megatron-lm-v25.6>`.
+
 .. 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 %}
+      {% for docker in dockers %}
       .. tab-item:: ``{{ docker.pull_tag }}``
          :sync: {{ docker.pull_tag }}
 
@@ -42,28 +56,14 @@ workloads:
 
             {% 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
    ================
 
-   The following models are supported for training performance benchmarking with Megatron-LM and ROCm.
+   The following models are supported for training performance benchmarking with Megatron-LM and ROCm
+   on AMD Instinct MI300X series accelerators.
    Some instructions, commands, and training recommendations in this documentation might
    vary by model -- select one to get started.
 
@@ -177,7 +177,7 @@ Download the Docker image
       {% if dockers|length > 1 %}
       .. tab-set::
 
-         {% for docker in data.dockers %}
+         {% for docker in dockers %}
          .. tab-item:: {{ docker.doc_name }}
             :sync: {{ docker.pull_tag }}
 
@@ -227,10 +227,17 @@ Download the Docker image
       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.
+4. **Megatron-LM backward compatibility setup** -- this Docker is primarily intended for use with Primus, but it maintains Megatron-LM compatibility with limited support.
+   To roll back to using Megatron-LM, follow these steps:
+
+   .. code-block:: shell
+
+      cd /workspace/Megatron-LM/
+      pip uninstall megatron-core
+      pip install -e .
+
+The Docker container hosts
+`<https://github.com/ROCm/Megatron-LM/tree/rocm_dev>`__ at verified commit ``e8e9edc``.
 
 .. _amd-megatron-lm-environment-setup:
 

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

@@ -16,12 +16,20 @@ previous releases of the ``ROCm/megatron-lm`` Docker image on `Docker Hub <https
      - Components
      - Resources
 
-   * - v25.6 (latest)
+   * - v25.7 (latest)
+     - 
+       * ROCm 
+       * PyTorch 
+     - 
+       * :doc:`Documentation <../megatron-lm>`
+       * `Docker Hub (py310) <https://hub.docker.com/layers/rocm/megatron-lm/v25.7_py310/images/sha256-6189df849feeeee3ae31bb1e97aef5006d69d2b90c134e97708c19632e20ab5a>`__
+
+   * - v25.6
      - 
        * ROCm 6.4.1
        * PyTorch 2.8.0a0+git7d205b2
      - 
-       * :doc:`Documentation <../megatron-lm>`
+       * :doc:`Documentation <megatron-lm-v25.6>`
        * `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>`__
 

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

@@ -0,0 +1,175 @@
+:orphan:
+
+**********************************************************************
+Migrating workloads to Primus (Megatron-Core backend) from Megatron-LM
+**********************************************************************
+
+Primus supports Megatron-Core as backend optimization library,
+replacing ROCm Megatron-LM. This document outlines the steps to migrate
+workload from ROCm Megatron-LM to Primus with the Megatron-Core backend.
+
+Model architecture
+==================
+
+ROCm Megatron-LM defines model architecture parameters in the training scripts;
+for example, the Llama 3 8B model parameters are defined in
+`examples/llama/train_llama3.sh <https://github.com/ROCm/Megatron-LM/blob/rocm_dev/examples/llama/train_llama3.sh#L117>`__
+as shown below:
+
+.. code-block:: bash
+
+   HIDDEN_SIZE=4096 
+   FFN_HIDDEN_SIZE=14336 
+   NUM_LAYERS=32 
+   NUM_HEADS=32 
+   NUM_KV_HEADS=8
+
+Primus defines the model architecture through model YAML configuration files
+inside the ``primus/configs/models/megatron/`` repository. For example, Llama 3 8B
+model architecture parameters are defined in
+`primus/configs/models/megatron/llama3_8B.yaml <https://github.com/AMD-AIG-AIMA/Primus/blob/v0.1.0-rc1/primus/configs/models/megatron/llama3_8B.yaml>`__
+as shown below:
+
+.. code-block:: yaml
+
+   bases:
+     - llama3_base.yaml
+
+   tokenizer_type: Llama3Tokenizer
+   tokenizer_model: meta-llama/Llama-3.1-8B
+
+   ffn_hidden_size: 14336
+   hidden_size: 4096
+   num_attention_heads: 32
+   num_layers: 32
+   num_query_groups: 8
+
+Primus' model config files follow a hierarchical design, meaning that new model
+config YAMLs can inherit existing model config files by importing them as
+bases. For example,
+`llama3.1_8B.yaml <https://github.com/AMD-AIG-AIMA/Primus/blob/v0.1.0-rc1/primus/configs/models/megatron/llama3.1_8B.yaml>`__
+uses ``llama3_8B.yaml`` as a base config and overrides few parameters, as shown below.
+In this example, ``llama3.1_8B`` overrides the ``max_position_embeddings`` value:
+
+.. code-block:: yaml
+
+   bases:
+     - llama3_8B.yaml
+
+   tokenizer_type: Llama3Tokenizer
+   tokenizer_model: meta-llama/Llama-3.1-8B
+
+   max_position_embeddings: 131072
+
+.. tip::
+
+   Primus provides ``llama_base.yaml`` as the base configuration, which can be
+   used as bases for additional model architectures. For example,
+   `mixtral_base.yaml <https://github.com/AMD-AIG-AIMA/Primus/blob/v0.1.0-rc1/primus/configs/models/megatron/mixtral_base.yaml>`__
+   and
+   `deepseek_v3_base.yaml <https://github.com/AMD-AIG-AIMA/Primus/blob/v0.1.0-rc1/primus/configs/models/megatron/deepseek_v3_base.yaml>`__
+   define ``llama_base.yaml`` as its base.
+
+   .. code-block:: yaml
+
+      # Example mixtral_base.yaml:
+
+      bases:
+        - llama_base.yaml
+
+      init_method_std: 0.01
+      rotary_base: 1000000
+      qk_layernorm: false
+
+      group_query_attention: true
+      num_query_groups: 8
+
+      # moe parameters
+      num_experts: 8
+      moe_router_topk: 2
+      moe_router_load_balancing_type: aux_loss
+      moe_aux_loss_coeff: 1e-2
+      moe_grouped_gemm: true
+      moe_token_dispatcher_type: alltoall
+
+It is recommended to add a new ``${MODEL_NAME}_base.yaml`` to add a new
+category of model and define new models on top of it. For example, to add
+Qwen2.5 models in Primus, we define
+`qwen2.5_base.yaml <https://github.com/AMD-AIG-AIMA/Primus/blob/v0.1.0-rc1/primus/configs/models/megatron/qwen2.5_base.yaml>`__
+and build
+`qwen2.5_7B.yaml <https://github.com/AMD-AIG-AIMA/Primus/blob/v0.1.0-rc1/primus/configs/models/megatron/qwen2.5_7B.yaml>`__
+and
+`qwen2.5_72B.yaml <https://github.com/AMD-AIG-AIMA/Primus/blob/v0.1.0-rc1/primus/configs/models/megatron/qwen2.5_72B.yaml>`__
+using ``qwen2.5_base.yaml`` as the base config.
+
+Training parameters
+===================
+
+ROCm Megatron-LM also defines the training parameters, like batch size,
+tensor-parallelism, precision, as so on, in the training scripts. For example,
+Llama3 8B model parameters are defined in
+`examples/llama/train_llama3.sh <https://github.com/ROCm/Megatron-LM/blob/rocm_dev/examples/llama/train_llama3.sh>`__
+as shown below:
+
+.. code-block:: bash
+
+   TP="${TP:-8}"
+   PP="${PP:-1}"
+   CP="${CP:-1}"
+   MBS="${MBS:-1}"
+   BS="${BS:-8}"
+
+Primus defines the training parameters in top-level YAML files -- see
+`examples/megatron/configs/
+<https://github.com/AMD-AIG-AIMA/Primus/tree/v0.1.0-rc1/examples/megatron/configs>`__.
+For example, the `llama3.1_8B-pretrain.yaml
+<https://github.com/AMD-AIG-AIMA/Primus/blob/v0.1.0-rc1/examples/megatron/configs/llama3.1_8B-pretrain.yaml>`__
+configuration imports the ``llama3.1_8B.yaml`` model architecture file. Users can then override
+the default training parameters in ``llama3.1_8B-pretrain.yaml``.
+
+.. code-block:: yaml
+
+   # model to run
+   model: llama3.1_8B.yaml  # Model architecture yaml
+   overrides:
+     # log
+     # disable_wandb: false
+     # disable_tensorboard: false
+     stderr_sink_level: DEBUG
+
+     log_avg_skip_iterations: 2
+     log_avg_reset_interval: 50
+
+     train_iters: 50
+     micro_batch_size: 2
+     global_batch_size: 128
+
+     seq_length: 8192
+     max_position_embeddings: 8192
+
+     lr: 1.0e-5
+     min_lr: 0.0
+     lr_warmup_iters: 2
+     lr_decay_iters: null
+     lr_decay_style: cosine
+     weight_decay: 0.1
+     adam_beta1: 0.9
+     adam_beta2: 0.95
+     eod_mask_loss: true
+     init_method_std: 0.008
+     norm_epsilon: 1.0e-6
+
+Backward compatibility with Megatron-LM
+=======================================
+
+The Dockerized environment used for Primus maintains compatibility with Megatron-LM with
+limited support. To roll back to using Megatron-LM, follow these steps.
+
+.. code-block:: shell
+
+   cd /workspace/Megatron-LM/
+   pip uninstall megatron-core
+   pip install -e .
+
+Once Megatron-LM is installed, follow :doc:`the documentation <../megatron-lm>` to run workloads as
+usual.

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

@@ -18,7 +18,7 @@ Training a model with ROCm Megatron-LM
 The ROCm Megatron-LM framework 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
 accelerators, AMD Megatron-LM delivers enhanced scalability, performance, and resource utilization for AI
-workloads. It is purpose-built to :ref:`support models <amd-megatron-lm-model-support>`
+workloads. It is purpose-built to :ref:`support models <amd-megatron-lm-model-support-24-12>`
 like Meta's Llama 2, Llama 3, and Llama 3.1, enabling developers to train next-generation AI models with greater
 efficiency. See the GitHub repository at `<https://github.com/ROCm/Megatron-LM>`__.
 
@@ -67,7 +67,7 @@ Megatron-LM provides the following key features to train large language models e
 
 - Pre-training
 
-.. _amd-megatron-lm-model-support:
+.. _amd-megatron-lm-model-support-24-12:
 
 The following models are pre-optimized for performance on the AMD Instinct MI300X accelerator.
 

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

@@ -67,7 +67,7 @@ Megatron-LM provides the following key features to train large language models e
 
 - Pre-training
 
-.. _amd-megatron-lm-model-support:
+.. _amd-megatron-lm-model-support-25-3:
 
 The following models are pre-optimized for performance on the AMD Instinct MI300X accelerator.
 
@@ -278,7 +278,7 @@ handle a variety of input sequences, including unseen words or domain-specific t
    .. tab-item:: Llama
       :sync: llama
 
-      To train any of the Llama 2 models that :ref:`this Docker image supports <amd-megatron-lm-model-support>`, use the ``Llama2Tokenizer``.
+      To train any of the Llama 2 models that :ref:`this Docker image supports <amd-megatron-lm-model-support-25-3>`, use the ``Llama2Tokenizer``.
 
       To train any of Llama 3 and Llama 3.1 models that this Docker image supports, use the ``HuggingFaceTokenizer``.
       Set the Hugging Face model link in the ``TOKENIZER_MODEL`` variable.
@@ -292,7 +292,7 @@ handle a variety of input sequences, including unseen words or domain-specific t
    .. tab-item:: DeepSeek V2
       :sync: deepseek
 
-      To train any of the DeepSeek V2 models that :ref:`this Docker image supports <amd-megatron-lm-model-support>`, use the ``DeepSeekV2Tokenizer``.
+      To train any of the DeepSeek V2 models that :ref:`this Docker image supports <amd-megatron-lm-model-support-25-3>`, use the ``DeepSeekV2Tokenizer``.
 
 Multi-node training
 ^^^^^^^^^^^^^^^^^^^

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

@@ -67,7 +67,7 @@ Megatron-LM provides the following key features to train large language models e
 
 - Pre-training
 
-.. _amd-megatron-lm-model-support:
+.. _amd-megatron-lm-model-support-25-4:
 
 The following models are pre-optimized for performance on AMD Instinct MI300X series accelerators.
 
@@ -291,7 +291,7 @@ or ``${DATA_DIR}/tokenizer_llama2``.
    .. tab-item:: Llama
       :sync: llama
 
-      To train any of the Llama 2 models that :ref:`this Docker image supports <amd-megatron-lm-model-support>`, use the ``Llama2Tokenizer``
+      To train any of the Llama 2 models that :ref:`this Docker image supports <amd-megatron-lm-model-support-25-4>`, use the ``Llama2Tokenizer``
       or the default ``HuggingFaceTokenizer``.
 
       To train any of Llama 3 and Llama 3.1 models that this Docker image supports, use the ``HuggingFaceTokenizer``.
@@ -320,7 +320,7 @@ or ``${DATA_DIR}/tokenizer_llama2``.
    .. tab-item:: DeepSeek V2
       :sync: deepseek
 
-      To train any of the DeepSeek V2 models that :ref:`this Docker image supports <amd-megatron-lm-model-support>`, use the ``DeepSeekV2Tokenizer``.
+      To train any of the DeepSeek V2 models that :ref:`this Docker image supports <amd-megatron-lm-model-support-25-4>`, use the ``DeepSeekV2Tokenizer``.
 
 Multi-node training
 ^^^^^^^^^^^^^^^^^^^

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

@@ -16,12 +16,20 @@ previous releases of the ``ROCm/pytorch-training`` Docker image on `Docker Hub <
      - Components
      - Resources
 
+   * - v25.7
+     - 
+       * ROCm 6.4.2
+       * PyTorch 2.8.0a0+gitd06a406
+     - 
+       * :doc:`Documentation <../pytorch-training>`
+       * `Docker Hub <https://hub.docker.com/layers/rocm/pytorch-training/v25.7/images/sha256-cc6fd840ab89cb81d926fc29eca6d075aee9875a55a522675a4b9231c9a0a712>`__
+
    * - v25.6
      - 
        * ROCm 6.3.4
        * PyTorch 2.8.0a0+git7d205b2
      - 
-       * :doc:`Documentation <../pytorch-training>`
+       * :doc:`Documentation <pytorch-training-v25.6>`
        * `Docker Hub <https://hub.docker.com/layers/rocm/pytorch-training/v25.6/images/sha256-a4cea3c493a4a03d199a3e81960ac071d79a4a7a391aa9866add3b30a7842661>`__
 
    * - v25.5

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

@@ -437,3 +437,8 @@ Once the setup is complete, choose between two options to start benchmarking:
 
            ./pytorch_benchmark_report.sh -t HF_finetune_lora -p BF16 -m Llama-2-70B
 
+Previous versions
+=================
+
+See :doc:`pytorch-training-history` to find documentation for previous releases
+of the ``ROCm/pytorch-training`` Docker image.

docs/how-to/rocm-for-ai/training/benchmark-docker/previous-versions/pytorch-training-v25.6.rst

@@ -0,0 +1,456 @@
+:orphan:
+
+.. meta::
+   :description: How to train a model using PyTorch for ROCm.
+   :keywords: ROCm, AI, LLM, train, PyTorch, torch, Llama, flux, tutorial, docker
+
+**************************************
+Training a model with PyTorch for ROCm
+**************************************
+
+.. caution::
+
+   This documentation does not reflect the latest version of ROCm vLLM
+   performance benchmark documentation. See :doc:`../pytorch-training` for the latest version.
+
+PyTorch is an open-source machine learning framework that is widely used for
+model training with GPU-optimized components for transformer-based models.
+
+The `PyTorch for ROCm training Docker <https://hub.docker.com/layers/rocm/pytorch-training/v25.6/images/sha256-a4cea3c493a4a03d199a3e81960ac071d79a4a7a391aa9866add3b30a7842661>`_
+(``rocm/pytorch-training:v25.6``) image provides a prebuilt optimized environment for fine-tuning and pretraining a
+model on AMD Instinct MI325X and MI300X accelerators. It includes the following software components to accelerate
+training workloads:
+
++--------------------------+--------------------------------+
+| Software component       | Version                        |
++==========================+================================+
+| ROCm                     | 6.3.4                          |
++--------------------------+--------------------------------+
+| PyTorch                  | 2.8.0a0+git7d205b2             |
++--------------------------+--------------------------------+
+| Python                   | 3.10.17                        |
++--------------------------+--------------------------------+
+| Transformer Engine       | 1.14.0+2f85f5f2                |
++--------------------------+--------------------------------+
+| Flash Attention          | 3.0.0.post1                    |
++--------------------------+--------------------------------+
+| hipBLASLt                | 0.15.0-8c6919d                 |
++--------------------------+--------------------------------+
+| Triton                   | 3.3.0                          |
++--------------------------+--------------------------------+
+
+.. _amd-pytorch-training-model-support-v256:
+
+Supported models
+================
+
+The following models are pre-optimized for performance on the AMD Instinct MI325X and MI300X accelerators.
+
+.. datatemplate:yaml:: /data/how-to/rocm-for-ai/training/previous-versions/pytorch-training-v25.6-benchmark-models.yaml
+
+   {% set unified_docker = data.unified_docker.latest %}
+   {% set model_groups = data.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">Workload</div>
+          <div class="row col-10">
+   {% for model_group in model_groups %}
+            <div class="col-6 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>
+
+   .. note::
+
+      Some models require an external license agreement through a third party (for example, Meta).
+
+   .. _amd-pytorch-training-performance-measurements-v256:
+
+   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.
+
+   .. note::
+
+      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#tabs-a8deaeb413-item-21cea50186-tab>`_
+      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.
+
+   This Docker image is optimized for specific model configurations outlined
+   below. Performance can vary for other training workloads, as AMD
+   doesn’t validate configurations and run conditions outside those described.
+
+   Benchmarking
+   ============
+
+   Once the setup is complete, choose between two options to start benchmarking:
+
+   .. 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
+
+   {% for model_group in model_groups %}
+      {% for model in model_group.models %}
+
+         .. container:: model-doc {{ model.mad_tag }}
+
+            For example, use this command to run the performance benchmark test on the {{ model.model }} model
+            using one GPU with the {{ model.precision }} data type on the host machine.
+
+            .. code-block:: shell
+
+               export MAD_SECRETS_HFTOKEN="your personal Hugging Face token to access gated models"
+               madengine run \
+                   --tags {{ model.mad_tag }} \
+                   --keep-model-dir \
+                   --live-output \
+                   --timeout 28800
+
+            MAD launches a Docker container with the name
+            ``container_ci-{{ model.mad_tag }}``, for example. The latency and throughput reports of the
+            model are collected in the following path: ``~/MAD/perf.csv``.
+
+      {% endfor %}
+   {% endfor %}
+
+      .. tab-item:: Standalone benchmarking
+
+         .. rubric:: Download the Docker image and required packages
+
+         Use the following command to pull the Docker image from Docker Hub.
+
+         .. code-block:: shell
+
+            docker pull {{ unified_docker.pull_tag }}
+
+         Run the Docker container.
+
+         .. code-block:: shell
+
+            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 $HOME:$HOME -v  $HOME/.ssh:/root/.ssh --shm-size 64G --name training_env {{ unified_docker.pull_tag }}
+
+         Use these commands if you exit the ``training_env`` container and need to return to it.
+
+         .. code-block:: shell
+
+            docker start training_env
+            docker exec -it training_env bash
+
+         In the Docker container, clone the `<https://github.com/ROCm/MAD>`__
+         repository and navigate to the benchmark scripts directory
+         ``/workspace/MAD/scripts/pytorch_train``.
+
+         .. code-block:: shell
+
+            git clone https://github.com/ROCm/MAD
+            cd MAD/scripts/pytorch_train
+
+         .. rubric:: Prepare training datasets and dependencies
+
+         The following benchmarking examples require downloading models and datasets
+         from Hugging Face. To ensure successful access to gated repos, set your
+         ``HF_TOKEN``.
+
+         .. code-block:: shell
+
+            export HF_TOKEN=$your_personal_hugging_face_access_token
+
+         Run the setup script to install libraries and datasets needed for benchmarking.
+
+         .. code-block:: shell
+
+            ./pytorch_benchmark_setup.sh
+
+         .. container:: model-doc pyt_train_llama-3.1-8b
+
+            ``pytorch_benchmark_setup.sh`` installs the following libraries for Llama 3.1 8B:
+
+            .. list-table::
+               :header-rows: 1
+
+               * - Library
+                 - Reference
+
+               * - ``accelerate``
+                 - `Hugging Face Accelerate <https://huggingface.co/docs/accelerate/en/index>`_
+
+               * - ``datasets``
+                 - `Hugging Face Datasets <https://huggingface.co/docs/datasets/v3.2.0/en/index>`_ 3.2.0
+
+         .. container:: model-doc pyt_train_llama-3.1-70b
+
+            ``pytorch_benchmark_setup.sh`` installs the following libraries for Llama 3.1 70B:
+
+            .. list-table::
+               :header-rows: 1
+
+               * - Library
+                 - Reference
+
+               * - ``datasets``
+                 - `Hugging Face Datasets <https://huggingface.co/docs/datasets/v3.2.0/en/index>`_ 3.2.0
+
+               * - ``torchdata``
+                 - `TorchData <https://pytorch.org/data/beta/index.html>`_
+
+               * - ``tomli``
+                 - `Tomli <https://pypi.org/project/tomli/>`_
+
+               * - ``tiktoken``
+                 - `tiktoken <https://github.com/openai/tiktoken>`_
+
+               * - ``blobfile``
+                 - `blobfile <https://pypi.org/project/blobfile/>`_
+
+               * - ``tabulate``
+                 - `tabulate <https://pypi.org/project/tabulate/>`_
+
+               * - ``wandb``
+                 - `Weights & Biases <https://github.com/wandb/wandb>`_
+
+               * - ``sentencepiece``
+                 - `SentencePiece <https://github.com/google/sentencepiece>`_ 0.2.0
+
+               * - ``tensorboard``
+                 - `TensorBoard <https://www.tensorflow.org/tensorboard>`_ 2.18.0
+
+         .. container:: model-doc pyt_train_flux
+
+            ``pytorch_benchmark_setup.sh`` installs the following libraries for FLUX:
+
+            .. list-table::
+               :header-rows: 1
+
+               * - Library
+                 - Reference
+
+               * - ``accelerate``
+                 - `Hugging Face Accelerate <https://huggingface.co/docs/accelerate/en/index>`_
+
+               * - ``datasets``
+                 - `Hugging Face Datasets <https://huggingface.co/docs/datasets/v3.2.0/en/index>`_ 3.2.0
+
+               * - ``sentencepiece``
+                 - `SentencePiece <https://github.com/google/sentencepiece>`_ 0.2.0
+
+               * - ``tensorboard``
+                 - `TensorBoard <https://www.tensorflow.org/tensorboard>`_ 2.18.0
+
+               * - ``csvkit``
+                 - `csvkit <https://csvkit.readthedocs.io/en/latest/>`_ 2.0.1
+
+               * - ``deepspeed``
+                 - `DeepSpeed <https://github.com/deepspeedai/DeepSpeed>`_ 0.16.2
+
+               * - ``diffusers``
+                 - `Hugging Face Diffusers <https://huggingface.co/docs/diffusers/en/index>`_ 0.31.0
+
+               * - ``GitPython``
+                 - `GitPython <https://github.com/gitpython-developers/GitPython>`_ 3.1.44
+
+               * - ``opencv-python-headless``
+                 - `opencv-python-headless <https://pypi.org/project/opencv-python-headless/>`_ 4.10.0.84
+
+               * - ``peft``
+                 - `PEFT <https://huggingface.co/docs/peft/en/index>`_ 0.14.0
+
+               * - ``protobuf``
+                 - `Protocol Buffers <https://github.com/protocolbuffers/protobuf>`_ 5.29.2
+
+               * - ``pytest``
+                 - `PyTest <https://docs.pytest.org/en/stable/>`_ 8.3.4
+
+               * - ``python-dotenv``
+                 - `python-dotenv <https://pypi.org/project/python-dotenv/>`_ 1.0.1
+
+               * - ``seaborn``
+                 - `Seaborn <https://seaborn.pydata.org/>`_ 0.13.2
+
+               * - ``transformers``
+                 - `Transformers <https://huggingface.co/docs/transformers/en/index>`_ 4.47.0
+
+         ``pytorch_benchmark_setup.sh`` downloads the following datasets from Hugging Face:
+
+         * `bghira/pseudo-camera-10k <https://huggingface.co/datasets/bghira/pseudo-camera-10k>`_
+
+   {% for model_group in model_groups %}
+      {% for model in model_group.models %}
+         {% if model_group.tag == "pre-training" and model.mad_tag in ["pyt_train_llama-3.1-8b", "pyt_train_llama-3.1-70b", "pyt_train_flux"] %}
+
+         .. container:: model-doc {{ model.mad_tag }}
+
+            .. rubric:: Pretraining
+
+            To start the pre-training benchmark, use the following command with the
+            appropriate options. See the following list of options and their descriptions.
+
+            .. code-block:: shell
+
+               ./pytorch_benchmark_report.sh -t pretrain -m {{ model.model_repo }} -p $datatype -s $sequence_length
+
+            .. list-table::
+               :header-rows: 1
+
+               * - Name
+                 - Options
+                 - Description
+
+            {% if model.mad_tag == "pyt_train_llama-3.1-8b" %}
+               * - ``$datatype``
+                 - ``BF16`` or ``FP8``
+                 - Only Llama 3.1 8B supports FP8 precision.
+            {% else %}
+               * - ``$datatype``
+                 - ``BF16``
+                 - Only Llama 3.1 8B supports FP8 precision.
+            {% endif %}
+
+               * - ``$sequence_length``
+                 - Sequence length for the language model.
+                 - Between 2048 and 8192. 8192 by default.
+
+            {% if model.mad_tag == "pyt_train_flux" %}
+            .. container:: model-doc {{ model.mad_tag }}
+
+               .. note::
+
+                  Occasionally, downloading the Flux dataset might fail. In the event of this
+                  error, manually download it from Hugging Face at
+                  `black-forest-labs/FLUX.1-dev <https://huggingface.co/black-forest-labs/FLUX.1-dev>`_
+                  and save it to `/workspace/FluxBenchmark`. This ensures that the test script can access
+                  the required dataset.
+            {% endif %}
+         {% endif %}
+
+         {% if model_group.tag == "fine-tuning" %}
+         .. container:: model-doc {{ model.mad_tag }}
+
+            .. rubric:: Fine-tuning
+
+            To start the fine-tuning benchmark, use the following command with the
+            appropriate options. See the following list of options and their descriptions.
+
+            .. code-block:: shell
+
+               ./pytorch_benchmark_report.sh -t $training_mode -m {{ model.model_repo }} -p BF16 -s $sequence_length
+
+            .. list-table::
+               :header-rows: 1
+
+               * - Name
+                 - Options
+                 - Description
+
+               * - ``$training_mode``
+                 - ``finetune_fw``
+                 - Full weight fine-tuning (BF16 supported)
+
+               * -
+                 - ``finetune_lora``
+                 - LoRA fine-tuning (BF16 supported)
+
+               * -
+                 - ``finetune_qlora``
+                 - QLoRA fine-tuning (BF16 supported)
+
+               * -
+                 - ``HF_finetune_lora``
+                 - LoRA fine-tuning with Hugging Face PEFT
+
+               * - ``$datatype``
+                 - ``BF16``
+                 - All models support BF16.
+
+               * - ``$sequence_length``
+                 - Between 2048 and 16384.
+                 - Sequence length for the language model.
+
+            .. note::
+
+               {{ model.model }} currently supports the following fine-tuning methods:
+
+            {% for method in model.training_modes %}
+               * ``{{ method }}``
+            {% endfor %}
+            {% if model.training_modes|length < 4 %}
+
+               The upstream `torchtune <https://github.com/pytorch/torchtune>`_ repository
+               does not currently provide YAML configuration files for other combinations of
+               model to fine-tuning method
+               However, you can still configure your own YAML files to enable support for
+               fine-tuning methods not listed here by following existing patterns in the
+               ``/workspace/torchtune/recipes/configs`` directory.
+            {% endif %}
+         {% endif %}
+      {% endfor %}
+   {% endfor %}
+
+               .. rubric:: Benchmarking examples
+
+               For examples of benchmarking commands, see `<https://github.com/ROCm/MAD/tree/develop/benchmark/pytorch_train#benchmarking-examples>`__.
+
+Further reading
+===============
+
+- To learn more about MAD and the ``madengine`` CLI, see the `MAD usage guide <https://github.com/ROCm/MAD?tab=readme-ov-file#usage-guide>`__.
+
+- To learn more about system settings and management practices to configure your system for
+  AMD Instinct MI300X series accelerators, see `AMD Instinct MI300X system optimization <https://instinct.docs.amd.com/projects/amdgpu-docs/en/latest/system-optimization/mi300x.html>`_.
+
+- 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:`pytorch-training-history` to find documentation for previous releases
+of the ``ROCm/pytorch-training`` Docker image.

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

@@ -0,0 +1,602 @@
+.. 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 Primus and Megatron-Core
+**********************************************
+
+`Primus <https://github.com/AMD-AIG-AIMA/Primus>`__ is a unified and flexible
+LLM training framework designed to streamline training. It streamlines LLM
+training on AMD Instinct accelerators using a modular, reproducible configuration paradigm.
+Primus is backend-agnostic and supports multiple training engines -- including Megatron-Core.
+
+.. note::
+
+   Primus with the Megatron-Core backend is intended to replace ROCm
+   Megatron-LM in this Dockerized training environment. To learn how to migrate
+   workloads from Megatron-LM to Primus with Megatron-Core, see
+   :doc:`previous-versions/megatron-lm-primus-migration-guide`.
+
+For ease of use, AMD provides a ready-to-use Docker image for MI300 series accelerators
+containing essential components for Primus and Megatron-Core.
+
+.. note::
+
+   This Docker environment is based on Python 3.10 and Ubuntu 22.04. For an alternative environment with
+   Python 3.12 and Ubuntu 24.04, see the :doc:`previous ROCm Megatron-LM v25.6 Docker release <previous-versions/megatron-lm-v25.6>`.
+
+.. datatemplate:yaml:: /data/how-to/rocm-for-ai/training/primus-megatron-benchmark-models.yaml
+
+   {% set dockers = data.dockers %}
+   {% set docker = dockers[0] %}
+   .. list-table::
+      :header-rows: 1
+
+      * - Software component
+        - Version
+
+      {% for component_name, component_version in docker.components.items() %}
+      * - {{ component_name }}
+        - {{ component_version }}
+      {% endfor %}
+
+.. _amd-primus-megatron-lm-model-support:
+
+Supported models
+================
+
+The following models are pre-optimized for performance on AMD Instinct MI300X series accelerators.
+Some instructions, commands, and training examples in this documentation might
+vary by model -- select one to get started.
+
+.. datatemplate:yaml:: /data/how-to/rocm-for-ai/training/primus-megatron-benchmark-models.yaml
+
+   {% set model_groups = data.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-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 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).
+
+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-primus-megatron-lm-training:
+
+.. datatemplate:yaml:: /data/how-to/rocm-for-ai/training/primus-megatron-benchmark-models.yaml
+
+   {% set dockers = data.dockers %}
+      {% set docker = dockers[0] %}
+
+   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 ``{{ docker.pull_tag }}`` image.
+
+   .. _amd-primus-megatron-lm-requirements:
+
+   Download the Docker image
+   -------------------------
+
+   1. Use the following command to pull the Docker image from Docker Hub.
+
+      .. code-block:: shell
+
+         docker pull {{ docker.pull_tag }}
+
+   2. Launch the Docker container.
+
+      .. 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 \
+             --shm-size 128G \
+             --name primus_training_env \
+             {{ docker.pull_tag }}
+
+3. Use these commands if you exit the ``primus_training_env`` container and need to return to it.
+
+   .. code-block:: shell
+
+      docker start primus_training_env
+      docker exec -it primus_training_env bash
+
+The Docker container hosts verified release tag ``v0.1.0-rc1`` of the `Primus
+<https://github.com/AMD-AIG-AIMA/Primus/tree/v0.1.0-rc1>`__ repository.
+
+.. _amd-primus-megatron-lm-environment-setup:
+
+Configuration
+=============
+
+Primus defines a training configuration in YAML for each model in
+`examples/megatron/configs <https://github.com/AMD-AIG-AIMA/Primus/tree/v0.1.0-rc1/examples/megatron/configs>`__.
+
+.. datatemplate:yaml:: /data/how-to/rocm-for-ai/training/primus-megatron-benchmark-models.yaml
+
+   {% set model_groups = data.model_groups %}
+   {% for model_group in model_groups %}
+      {% for model in model_group.models %}
+   .. container:: model-doc {{ model.mad_tag }}
+
+      To update training parameters for {{ model.model }}, you can update ``examples/megatron/configs/{{ model.config_name }}``.
+      Note that training configuration YAML files for other models follow this naming convention.
+
+      {% endfor %}
+   {% endfor %}
+
+.. note::
+
+   See :ref:`Key options <amd-primus-megatron-lm-benchmark-test-vars>` for more information on configuration options.
+
+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`` field to toggle between mock and real data. The default
+  value is ``true`` for enabled.
+
+  .. code-block:: yaml
+
+     mock_data: true
+
+* If you're using a real dataset, update the ``train_data_path`` field to point to the location of your dataset.
+
+  .. code-block:: bash
+
+     mock_data: false
+     train_data_path: /path/to/your/dataset
+
+  Ensure that the files are accessible inside the Docker container.
+
+.. _amd-primus-megatron-lm-tokenizer:
+
+Tokenizer
+---------
+
+In Primus, each model uses a tokenizer from Hugging Face. For example, Llama
+3.1 8B model uses ``tokenizer_model: meta-llama/Llama-3.1-8B`` and
+``tokenizer_type: Llama3Tokenizer`` defined in the `llama3.1-8B model
+<https://github.com/AMD-AIG-AIMA/Primus/tree/v0.1.0-rc1/primus/configs/models/megatron/llama3.1_8B.yaml>`__
+definition. As such, you need to set the ``HF_TOKEN`` environment variable with
+right permissions to access the tokenizer for each model.
+
+.. code-block:: bash
+
+   # Export your HF_TOKEN in the workspace
+   export HF_TOKEN=<your_hftoken>
+
+.. _amd-primus-megatron-lm-run-training:
+
+Run training
+============
+
+Use the following example commands to set up the environment, configure
+:ref:`key options <amd-primus-megatron-lm-benchmark-test-vars>`, and run training on
+MI300X series accelerators with the AMD Megatron-LM environment.
+
+Single node training
+--------------------
+
+To run training on a single node, navigate to ``/workspace/Primus`` and use the following setup command:
+
+.. code-block:: shell
+
+   pip install -r requirements.txt
+   export HSA_NO_SCRATCH_RECLAIM=1
+   export NVTE_CK_USES_BWD_V3=1
+
+Once setup is complete, run the appropriate training command.
+
+.. container:: model-doc primus_pyt_megatron_lm_train_llama-3.3-70b
+
+   To run pre-training for Llama 3.3 70B BF16, run:
+
+   .. code-block:: shell
+
+      EXP=examples/megatron/configs/llama3.3_70B-pretrain.yaml \
+      bash ./examples/run_pretrain.sh \
+          --micro_batch_size 2 \
+          --global_batch_size 16 \
+          --train_iters 50
+
+.. container:: model-doc primus_pyt_megatron_lm_train_llama-3.1-8b
+
+   To run pre-training for Llama 3.1 8B FP8, run:
+
+   .. code-block:: shell
+
+      EXP=examples/megatron/configs/llama3.1_8B-pretrain.yaml \
+      bash ./examples/run_pretrain.sh \
+          --train_iters 50 \
+          --fp8 hybrid
+
+   For Llama 3.1 8B BF16, use the following command:
+
+   .. code-block:: shell
+
+      EXP=examples/megatron/configs/llama3.1_8B-pretrain.yaml \
+      bash ./examples/run_pretrain.sh --train_iters 50
+
+.. container:: model-doc primus_pyt_megatron_lm_train_llama-3.1-70b
+
+   To run pre-training for Llama 3.1 70B BF16, run:
+
+   .. code-block:: shell
+
+      EXP=examples/megatron/configs/llama3.1_70B-pretrain.yaml \
+      bash ./examples/run_pretrain.sh \
+           --train_iters 50
+
+   To run the training on a single node for Llama 3.1 70B FP8 with proxy, use the following command:
+
+   .. code-block:: shell
+
+      EXP=examples/megatron/configs/llama3.1_70B-pretrain.yaml \
+      bash ./examples/run_pretrain.sh \
+          --train_iters 50 \
+          --num_layers 40 \
+          --fp8 hybrid \
+          --no_fp8_weight_transpose_cache true
+
+   .. note::
+
+      Use two or more nodes to run the *full* Llama 70B model with FP8 precision.
+
+.. container:: model-doc primus_pyt_megatron_lm_train_llama-2-7b
+
+   To run pre-training for Llama 2 7B FP8, run:
+
+   .. code-block:: shell
+
+      EXP=examples/megatron/configs/llama2_7B-pretrain.yaml \
+      bash ./examples/run_pretrain.sh \
+          --train_iters 50 \
+          --fp8 hybrid
+
+   To run pre-training for Llama 2 7B BF16, run:
+
+   .. code-block:: shell
+
+      EXP=examples/megatron/configs/llama2_7B-pretrain.yaml \
+      bash ./examples/run_pretrain.sh --train_iters 50
+
+.. container:: model-doc primus_pyt_megatron_lm_train_llama-2-70b
+
+   To run pre-training for Llama 2 70B BF16, run:
+
+   .. code-block:: shell
+
+      EXP=examples/megatron/configs/llama2_70B-pretrain.yaml \
+      bash ./examples/run_pretrain.sh --train_iters 50 
+
+.. container:: model-doc primus_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, 
+   use the following command:
+
+   .. code-block:: shell
+
+      EXP=examples/megatron/configs/deepseek_v3-pretrain.yaml \
+      bash examples/run_pretrain.sh \
+          --num_layers 3 \
+          --moe_layer_freq 1 \
+          --train_iters 50
+
+.. container:: model-doc primus_pyt_megatron_lm_train_deepseek-v2-lite-16b
+
+   To run training on a single node for DeepSeek-V2-Lite (MoE with expert parallel),
+   use the following command:
+
+   .. code-block:: shell
+
+      EXP=examples/megatron/configs/deepseek_v2_lite-pretrain.yaml \
+      bash examples/run_pretrain.sh \
+          --global_batch_size 256 \
+          --train_iters 50
+
+.. container:: model-doc primus_pyt_megatron_lm_train_mixtral-8x7b
+
+   To run training on a single node for Mixtral 8x7B (MoE with expert parallel),
+   use the following command:
+
+   .. code-block:: shell
+
+      EXP=examples/megatron/configs/mixtral_8x7B_v0.1-pretrain.yaml \
+      bash examples/run_pretrain.sh --train_iters 50
+
+.. container:: model-doc primus_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,
+   use the following command:
+
+   .. code-block:: shell
+
+      EXP=examples/megatron/configs/mixtral_8x22B_v0.1-pretrain.yaml \
+      bash examples/run_pretrain.sh \
+          --num_layers 4 \
+          --pipeline_model_parallel_size 1 \
+          --micro_batch_size 1 \
+          --global_batch_size 16 \
+          --train_iters 50
+
+.. container:: model-doc primus_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
+
+      EXP=examples/megatron/configs/qwen2.5_7B-pretrain.yaml \
+      bash examples/run_pretrain.sh --train_iters 50
+
+   For FP8, use the following command.
+
+   .. code-block:: shell
+
+      EXP=examples/megatron/configs/qwen2.5_7B-pretrain.yaml \
+      bash examples/run_pretrain.sh \
+          --train_iters 50 \
+          --fp8 hybrid
+
+.. container:: model-doc primus_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
+
+      EXP=examples/megatron/configs/qwen2.5_72B-pretrain.yaml \
+      bash examples/run_pretrain.sh --train_iters 50
+
+Multi-node training examples
+----------------------------
+
+To run training on multiple nodes, you can use the
+`run_slurm_pretrain.sh <https://github.com/AMD-AIG-AIMA/Primus/tree/v0.1.0-rc1/examples/run_slurm_pretrain.sh>`__
+to launch the multi-node workload. Use the following steps to setup your environment:
+
+.. datatemplate:yaml:: /data/how-to/rocm-for-ai/training/primus-megatron-benchmark-models.yaml
+
+   {% set dockers = data.dockers %}
+   {% set docker = dockers[0] %}
+
+   .. code-block:: shell
+
+      cd /workspace/Primus/
+      export DOCKER_IMAGE={{ docker.pull_tag }}
+      export HF_TOKEN=<your_HF_token>
+      export HSA_NO_SCRATCH_RECLAIM=1
+      export NVTE_CK_USES_BWD_V3=1
+      export NCCL_IB_HCA=<your_NCCL_IB_HCA> # specify which RDMA interfaces to use for communication
+      export NCCL_SOCKET_IFNAME=<your_NCCL_SOCKET_IFNAME> # your Network Interface
+      export GLOO_SOCKET_IFNAME=<your_GLOO_SOCKET_IFNAME> # your Network Interface
+      export NCCL_IB_GID_INDEX=3 # Set InfiniBand GID index for NCCL communication. Default is 3 for ROCE
+
+.. note::
+
+   * Make sure correct network drivers are installed on the nodes. If inside a Docker, either install the drivers inside the Docker container or pass the network drivers from the host while creating Docker container.
+   * If ``NCCL_IB_HCA`` and ``NCCL_SOCKET_IFNAME`` are not set, Primus will try to auto-detect. However, since NICs can vary accross different cluster, it is encouraged to explicitly export your NCCL parameters for the cluster.
+   * To find your network interface, you can use ``ip a``.
+   * To find RDMA interfaces, you can use ``ibv_devices`` to get the list of all the RDMA/IB  devices.
+
+.. container:: model-doc primus_pyt_megatron_lm_train_llama-3.3-70b
+
+   To train Llama 3.3 70B FP8 on 8 nodes, run:
+
+   .. code-block:: shell
+
+      NNODES=8 EXP=examples/megatron/configs/llama3.3_70B-pretrain.yaml \
+      bash examples/run_slurm_pretrain.sh \
+          --micro_batch_size 4 \
+          --global_batch_size 256 \
+          --recompute_num_layers 80 \
+          --no_fp8_weight_transpose_cache true \
+          --fp8 hybrid
+
+   To train Llama 3.3 70B BF16 on 8 nodes, run:
+
+   .. code-block:: shell
+
+      NNODES=8 EXP=examples/megatron/configs/llama3.3_70B-pretrain.yaml \
+      bash examples/run_slurm_pretrain.sh \
+          --micro_batch_size 1 \
+          --global_batch_size 256 \
+          --recompute_num_layers 12
+
+.. container:: model-doc primus_pyt_megatron_lm_train_llama-3.1-8b
+
+   To train Llama 3.1 8B FP8 on 8 nodes, run:
+
+   .. code-block:: shell
+
+      # Adjust the training parameters. For e.g., `global_batch_size: 8 * #single_node_bs` for 8 nodes in this case 
+      NNODES=8 EXP=examples/megatron/configs/llama3.1_8B-pretrain.yaml \
+      bash ./examples/run_slurm_pretrain.sh \
+          --global_batch_size 1024 \
+          --fp8 hybrid
+
+.. container:: model-doc primus_pyt_megatron_lm_train_llama-3.1-70b
+
+   To train Llama 3.1 70B FP8 on 8 nodes, run:
+
+   .. code-block:: shell
+
+      NNODES=8 EXP=examples/megatron/configs/llama3.1_70B-pretrain.yaml \
+      bash examples/run_slurm_pretrain.sh \
+          --micro_batch_size 4 \
+          --global_batch_size 256 \
+          --recompute_num_layers 80 \
+          --no_fp8_weight_transpose_cache true \
+          --fp8 hybrid
+
+   To train Llama 3.1 70B BF16 on 8 nodes, run:
+
+   .. code-block:: shell
+
+      NNODES=8 EXP=examples/megatron/configs/llama3.1_70B-pretrain.yaml \
+      bash examples/run_slurm_pretrain.sh \
+          --micro_batch_size 1 \
+          --global_batch_size 256 \
+          --recompute_num_layers 12
+
+.. container:: model-doc primus_pyt_megatron_lm_train_llama-2-7b
+
+   To train Llama 2 8B FP8 on 8 nodes, run:
+
+   .. code-block:: shell
+
+      # Adjust the training parameters. For e.g., `global_batch_size: 8 * #single_node_bs` for 8 nodes in this case 
+      NNODES=8 EXP=examples/megatron/configs/llama2_7B-pretrain.yaml bash ./examples/run_slurm_pretrain.sh --global_batch_size 2048 --fp8 hybrid
+
+.. container:: model-doc primus_pyt_megatron_lm_train_llama-2-70b
+
+   To train Llama 2 70B FP8 on 8 nodes, run:
+
+   .. code-block:: shell
+
+      NNODES=8 EXP=examples/megatron/configs/llama2_70B-pretrain.yaml \
+      bash examples/run_slurm_pretrain.sh \
+          --micro_batch_size 10 \
+          --global_batch_size 640 \
+          --recompute_num_layers 80 \
+          --no_fp8_weight_transpose_cache true \
+          --fp8 hybrid
+
+   To train Llama 2 70B BF16 on 8 nodes, run:
+
+   .. code-block:: shell
+
+      NNODES=8 EXP=examples/megatron/configs/llama2_70B-pretrain.yaml \
+      bash ./examples/run_slurm_pretrain.sh \
+          --micro_batch_size 2 \
+          --global_batch_size 1536 \
+          --recompute_num_layers 12
+
+.. container:: model-doc primus_pyt_megatron_lm_train_mixtral-8x7b
+
+   To train Mixtral 8x7B BF16 on 8 nodes, run:
+
+   .. code-block:: shell
+
+      NNODES=8 EXP=examples/megatron/configs/mixtral_8x7B_v0.1-pretrain.yaml \
+      bash examples/run_slurm_pretrain.sh \
+          --micro_batch_size 2 \
+          --global_batch_size 256
+
+.. container:: model-doc primus_pyt_megatron_lm_train_qwen2.5-72b
+
+   To train Qwen2.5 72B FP8 on 8 nodes, run:
+
+   .. code-block:: shell
+
+      NNODES=8 EXP=examples/megatron/configs/qwen2.5_72B-pretrain.yaml \
+      bash examples/run_slurm_pretrain.sh \
+          --micro_batch_size 8 \
+          --global_batch_size 512 \
+          --recompute_num_layers 80 \
+          --no_fp8_weight_transpose_cache true \
+          --fp8 hybrid
+
+.. _amd-primus-megatron-lm-benchmark-test-vars:
+
+Key options
+-----------
+
+The following are key options to take note of
+
+fp8
+  ``hybrid`` enables FP8 GEMMs.
+
+use_torch_fsdp2
+  ``use_torch_fsdp2: 1``  enables torch fsdp-v2. If FSDP is enabled,
+  set ``use_distributed_optimizer`` and ``overlap_param_gather`` to ``false``.
+
+profile
+  To enable PyTorch profiling, set these parameters:
+
+  .. code-block:: yaml
+
+     profile: true
+     use_pytorch_profiler: true
+     profile_step_end: 7
+     profile_step_start: 6
+
+train_iters
+  The total number of iterations (default: 50).
+
+mock_data
+  True by default.
+
+micro_batch_size
+  Micro batch size.
+
+global_batch_size
+  Global batch size.
+
+recompute_granularity
+  For activation checkpointing.
+
+num_layers
+  For using a reduced number of layers as with proxy models.
+
+Previous versions
+=================
+
+See :doc:`previous-versions/megatron-lm-history` to find documentation for previous releases
+of the ``ROCm/megatron-lm`` Docker image.
+
+This training environment now uses Primus with Megatron as the primary
+configuration. Limited support for the legacy ROCm Megatron-LM is still
+available. For instructions on using ROCm Megatron-LM, see the
+:doc:`megatron-lm` document.

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

@@ -9,28 +9,25 @@ Training a model with PyTorch for ROCm
 PyTorch is an open-source machine learning framework that is widely used for
 model training with GPU-optimized components for transformer-based models.
 
-The `PyTorch for ROCm training Docker <https://hub.docker.com/r/rocm/pytorch-training/tags>`_
-(``rocm/pytorch-training:v25.6``) image provides a prebuilt optimized environment for fine-tuning and pretraining a
-model on AMD Instinct MI325X and MI300X accelerators. It includes the following software components to accelerate
-training workloads:
+.. datatemplate:yaml:: /data/how-to/rocm-for-ai/training/pytorch-training-benchmark-models.yaml
 
-+--------------------------+--------------------------------+
-| Software component       | Version                        |
-+==========================+================================+
-| ROCm                     | 6.3.4                          |
-+--------------------------+--------------------------------+
-| PyTorch                  | 2.8.0a0+git7d205b2             |
-+--------------------------+--------------------------------+
-| Python                   | 3.10.17                        |
-+--------------------------+--------------------------------+
-| Transformer Engine       | 1.14.0+2f85f5f2                |
-+--------------------------+--------------------------------+
-| Flash Attention          | 3.0.0.post1                    |
-+--------------------------+--------------------------------+
-| hipBLASLt                | 0.15.0-8c6919d                 |
-+--------------------------+--------------------------------+
-| Triton                   | 3.3.0                          |
-+--------------------------+--------------------------------+
+   {% set dockers = data.dockers %}
+   {% set docker = dockers[0] %}
+   The `PyTorch for ROCm training Docker <{{ docker.docker_hub_url }}>`__
+   (``{{ docker.pull_tag }}``) image provides a prebuilt optimized environment for fine-tuning and pretraining a
+   model on AMD Instinct MI325X and MI300X accelerators. It includes the following software components to accelerate
+   training workloads:
+
+   .. list-table::
+      :header-rows: 1
+
+      * - Software component
+        - Version
+
+      {% for component_name, component_version in docker.components.items() %}
+      * - {{ component_name }}
+        - {{ component_version }}
+      {% endfor %}
 
 .. _amd-pytorch-training-model-support:
 
@@ -38,26 +35,27 @@ Supported models
 ================
 
 The following models are pre-optimized for performance on the AMD Instinct MI325X and MI300X accelerators.
+Some instructions, commands, and training recommendations in this documentation might
+vary by model -- select one to get started.
 
 .. datatemplate:yaml:: /data/how-to/rocm-for-ai/training/pytorch-training-benchmark-models.yaml
 
-   {% set unified_docker = data.unified_docker.latest %}
+   {% set unified_docker = data.dockers[0] %}
    {% set model_groups = data.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">Workload</div>
+          <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-6 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>
 
         <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 variant</div>
           <div class="row col-10">
    {% for model_group in model_groups %}
       {% set models = model_group.models %}
@@ -73,84 +71,116 @@ The following models are pre-optimized for performance on the AMD Instinct MI325
         </div>
       </div>
 
-   .. note::
 
-      Some models require an external license agreement through a third party (for example, Meta).
+   .. _amd-pytorch-training-supported-training-modes:
 
-   .. _amd-pytorch-training-performance-measurements:
+   The following table lists supported training modes per model.
 
-   Performance measurements
-   ========================
+   .. dropdown:: Supported training modes
 
-   To evaluate performance, the
+      .. list-table::
+         :header-rows: 1
+
+         * - Model
+           - Supported training modes
+
+      {% for model_group in model_groups %}
+         {% set models = model_group.models %}
+         {% for model in models %}
+         * - {{ model.model }}
+           - ``{{ model.training_modes | join('``, ``') }}``
+
+         {% endfor %}
+      {% endfor %}
+
+      .. note::
+
+         Some model and fine-tuning combinations are not listed. This is
+         because the `upstream torchtune repository <https://github.com/pytorch/torchtune>`__
+         doesn't provide default YAML configurations for them.
+         For advanced usage, you can create a custom configuration to enable
+         unlisted fine-tuning methods by using an existing file in the
+         ``/workspace/torchtune/recipes/configs`` directory as a template.
+
+.. _amd-pytorch-training-performance-measurements:
+
+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.
+
+.. note::
+
+   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#tabs-a8deaeb413-item-21cea50186-tab>`_
-   page provides reference throughput and latency measurements for training
-   popular AI models.
+   should not be interpreted as the peak performance achievable by AMD
+   Instinct MI325X and MI300X accelerators or ROCm software.
 
-   .. note::
+System validation
+=================
 
-      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#tabs-a8deaeb413-item-21cea50186-tab>`_
-      should not be interpreted as the peak performance achievable by AMD
-      Instinct MI325X and MI300X accelerators or ROCm software.
+Before running AI workloads, it's important to validate that your AMD hardware is configured
+correctly and performing optimally.
 
-   System validation
-   =================
+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.
 
-   Before running AI workloads, it's important to validate that your AMD hardware is configured
-   correctly and performing optimally.
+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.
 
-   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.
+This Docker image is optimized for specific model configurations outlined
+below. Performance can vary for other training workloads, as AMD
+doesn’t test configurations and run conditions outside those described.
 
-   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.
+Run training
+============
 
-   This Docker image is optimized for specific model configurations outlined
-   below. Performance can vary for other training workloads, as AMD
-   doesn’t validate configurations and run conditions outside those described.
+.. datatemplate:yaml:: /data/how-to/rocm-for-ai/training/pytorch-training-benchmark-models.yaml
 
-   Benchmarking
-   ============
+   {% set unified_docker = data.dockers[0] %}
+   {% set model_groups = data.model_groups %}
 
-   Once the setup is complete, choose between two options to start benchmarking:
+   Once the setup is complete, choose between two options to start benchmarking training:
 
    .. 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.
+         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.
 
-         .. code-block:: shell
+            .. code-block:: shell
 
-            git clone https://github.com/ROCm/MAD
-            cd MAD
-            pip install -r requirements.txt
+               git clone https://github.com/ROCm/MAD
+               cd MAD
+               pip install -r requirements.txt
 
    {% for model_group in model_groups %}
       {% for model in model_group.models %}
 
          .. container:: model-doc {{ model.mad_tag }}
 
-            For example, use this command to run the performance benchmark test on the {{ model.model }} model
-            using one GPU with the {{ model.precision }} data type on the host machine.
+            2. For example, use this command to run the performance benchmark test on the {{ model.model }} model
+               using one node with the {{ 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"
-               madengine run \
-                   --tags {{ model.mad_tag }} \
-                   --keep-model-dir \
-                   --live-output \
-                   --timeout 28800
+                  export MAD_SECRETS_HFTOKEN="your personal Hugging Face token to access gated models"
+                  madengine run \
+                      --tags {{ model.mad_tag }} \
+                      --keep-model-dir \
+                      --live-output \
+                      --timeout 28800
 
-            MAD launches a Docker container with the name
-            ``container_ci-{{ model.mad_tag }}``, for example. The latency and throughput reports of the
-            model are collected in the following path: ``~/MAD/perf.csv``.
+               MAD launches a Docker container with the name
+               ``container_ci-{{ model.mad_tag }}``. The latency and throughput reports of the
+               model are collected in ``~/MAD/perf.csv``.
 
       {% endfor %}
    {% endfor %}
@@ -159,222 +189,213 @@ The following models are pre-optimized for performance on the AMD Instinct MI325
 
          .. rubric:: Download the Docker image and required packages
 
-         Use the following command to pull the Docker image from Docker Hub.
+         1. Use the following command to pull the Docker image from Docker Hub.
 
-         .. code-block:: shell
+            .. code-block:: shell
 
-            docker pull {{ unified_docker.pull_tag }}
+               docker pull {{ unified_docker.pull_tag }}
 
-         Run the Docker container.
+         2. Run the Docker container.
 
-         .. code-block:: shell
+            .. code-block:: shell
 
-            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 $HOME:$HOME -v  $HOME/.ssh:/root/.ssh --shm-size 64G --name training_env {{ unified_docker.pull_tag }}
+               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 $HOME:$HOME \
+                   -v $HOME/.ssh:/root/.ssh \
+                   --shm-size 64G \
+                   --name training_env \
+                   {{ unified_docker.pull_tag }}
 
-         Use these commands if you exit the ``training_env`` container and need to return to it.
+            Use these commands if you exit the ``training_env`` container and need to return to it.
 
-         .. code-block:: shell
+            .. code-block:: shell
 
-            docker start training_env
-            docker exec -it training_env bash
+               docker start training_env
+               docker exec -it training_env bash
 
-         In the Docker container, clone the `<https://github.com/ROCm/MAD>`__
-         repository and navigate to the benchmark scripts directory
-         ``/workspace/MAD/scripts/pytorch_train``.
+         3. In the Docker container, clone the `<https://github.com/ROCm/MAD>`__
+            repository and navigate to the benchmark scripts directory
+            ``/workspace/MAD/scripts/pytorch_train``.
 
-         .. code-block:: shell
+            .. code-block:: shell
 
-            git clone https://github.com/ROCm/MAD
-            cd MAD/scripts/pytorch_train
+               git clone https://github.com/ROCm/MAD
+               cd MAD/scripts/pytorch_train
 
          .. rubric:: Prepare training datasets and dependencies
 
-         The following benchmarking examples require downloading models and datasets
-         from Hugging Face. To ensure successful access to gated repos, set your
-         ``HF_TOKEN``.
+         1. The following benchmarking examples require downloading models and datasets
+            from Hugging Face. To ensure successful access to gated repos, set your
+            ``HF_TOKEN``.
 
-         .. code-block:: shell
+            .. code-block:: shell
 
-            export HF_TOKEN=$your_personal_hugging_face_access_token
+               export HF_TOKEN=$your_personal_hugging_face_access_token
 
-         Run the setup script to install libraries and datasets needed for benchmarking.
+         2. Run the setup script to install libraries and datasets needed for benchmarking.
 
-         .. code-block:: shell
+            .. code-block:: shell
 
-            ./pytorch_benchmark_setup.sh
+               ./pytorch_benchmark_setup.sh
 
-         .. container:: model-doc pyt_train_llama-3.1-8b
+            .. container:: model-doc pyt_train_llama-3.1-8b
 
-            ``pytorch_benchmark_setup.sh`` installs the following libraries for Llama 3.1 8B:
+               ``pytorch_benchmark_setup.sh`` installs the following libraries for Llama 3.1 8B:
 
-            .. list-table::
-               :header-rows: 1
+               .. list-table::
+                  :header-rows: 1
 
-               * - Library
-                 - Reference
+                  * - Library
+                    - Reference
 
-               * - ``accelerate``
-                 - `Hugging Face Accelerate <https://huggingface.co/docs/accelerate/en/index>`_
+                  * - ``accelerate``
+                    - `Hugging Face Accelerate <https://huggingface.co/docs/accelerate/en/index>`_
 
-               * - ``datasets``
-                 - `Hugging Face Datasets <https://huggingface.co/docs/datasets/v3.2.0/en/index>`_ 3.2.0
+                  * - ``datasets``
+                    - `Hugging Face Datasets <https://huggingface.co/docs/datasets/v3.2.0/en/index>`_ 3.2.0
 
-         .. container:: model-doc pyt_train_llama-3.1-70b
+            .. container:: model-doc pyt_train_llama-3.1-70b
 
-            ``pytorch_benchmark_setup.sh`` installs the following libraries for Llama 3.1 70B:
+               ``pytorch_benchmark_setup.sh`` installs the following libraries for Llama 3.1 70B:
 
-            .. list-table::
-               :header-rows: 1
+               .. list-table::
+                  :header-rows: 1
 
-               * - Library
-                 - Reference
+                  * - Library
+                    - Reference
 
-               * - ``datasets``
-                 - `Hugging Face Datasets <https://huggingface.co/docs/datasets/v3.2.0/en/index>`_ 3.2.0
+                  * - ``datasets``
+                    - `Hugging Face Datasets <https://huggingface.co/docs/datasets/v3.2.0/en/index>`_ 3.2.0
 
-               * - ``torchdata``
-                 - `TorchData <https://pytorch.org/data/beta/index.html>`_
+                  * - ``torchdata``
+                    - `TorchData <https://pytorch.org/data/beta/index.html>`_
 
-               * - ``tomli``
-                 - `Tomli <https://pypi.org/project/tomli/>`_
+                  * - ``tomli``
+                    - `Tomli <https://pypi.org/project/tomli/>`_
 
-               * - ``tiktoken``
-                 - `tiktoken <https://github.com/openai/tiktoken>`_
+                  * - ``tiktoken``
+                    - `tiktoken <https://github.com/openai/tiktoken>`_
 
-               * - ``blobfile``
-                 - `blobfile <https://pypi.org/project/blobfile/>`_
+                  * - ``blobfile``
+                    - `blobfile <https://pypi.org/project/blobfile/>`_
 
-               * - ``tabulate``
-                 - `tabulate <https://pypi.org/project/tabulate/>`_
+                  * - ``tabulate``
+                    - `tabulate <https://pypi.org/project/tabulate/>`_
 
-               * - ``wandb``
-                 - `Weights & Biases <https://github.com/wandb/wandb>`_
+                  * - ``wandb``
+                    - `Weights & Biases <https://github.com/wandb/wandb>`_
 
-               * - ``sentencepiece``
-                 - `SentencePiece <https://github.com/google/sentencepiece>`_ 0.2.0
+                  * - ``sentencepiece``
+                    - `SentencePiece <https://github.com/google/sentencepiece>`_ 0.2.0
 
-               * - ``tensorboard``
-                 - `TensorBoard <https://www.tensorflow.org/tensorboard>`_ 2.18.0
+                  * - ``tensorboard``
+                    - `TensorBoard <https://www.tensorflow.org/tensorboard>`_ 2.18.0
 
-         .. container:: model-doc pyt_train_flux
+            .. container:: model-doc pyt_train_flux
 
-            ``pytorch_benchmark_setup.sh`` installs the following libraries for FLUX:
+               ``pytorch_benchmark_setup.sh`` installs the following libraries for FLUX:
 
-            .. list-table::
-               :header-rows: 1
+               .. list-table::
+                  :header-rows: 1
 
-               * - Library
-                 - Reference
+                  * - Library
+                    - Reference
 
-               * - ``accelerate``
-                 - `Hugging Face Accelerate <https://huggingface.co/docs/accelerate/en/index>`_
+                  * - ``accelerate``
+                    - `Hugging Face Accelerate <https://huggingface.co/docs/accelerate/en/index>`_
 
-               * - ``datasets``
-                 - `Hugging Face Datasets <https://huggingface.co/docs/datasets/v3.2.0/en/index>`_ 3.2.0
+                  * - ``datasets``
+                    - `Hugging Face Datasets <https://huggingface.co/docs/datasets/v3.2.0/en/index>`_ 3.2.0
 
-               * - ``sentencepiece``
-                 - `SentencePiece <https://github.com/google/sentencepiece>`_ 0.2.0
+                  * - ``sentencepiece``
+                    - `SentencePiece <https://github.com/google/sentencepiece>`_ 0.2.0
 
-               * - ``tensorboard``
-                 - `TensorBoard <https://www.tensorflow.org/tensorboard>`_ 2.18.0
+                  * - ``tensorboard``
+                    - `TensorBoard <https://www.tensorflow.org/tensorboard>`_ 2.18.0
 
-               * - ``csvkit``
-                 - `csvkit <https://csvkit.readthedocs.io/en/latest/>`_ 2.0.1
+                  * - ``csvkit``
+                    - `csvkit <https://csvkit.readthedocs.io/en/latest/>`_ 2.0.1
 
-               * - ``deepspeed``
-                 - `DeepSpeed <https://github.com/deepspeedai/DeepSpeed>`_ 0.16.2
+                  * - ``deepspeed``
+                    - `DeepSpeed <https://github.com/deepspeedai/DeepSpeed>`_ 0.16.2
 
-               * - ``diffusers``
-                 - `Hugging Face Diffusers <https://huggingface.co/docs/diffusers/en/index>`_ 0.31.0
+                  * - ``diffusers``
+                    - `Hugging Face Diffusers <https://huggingface.co/docs/diffusers/en/index>`_ 0.31.0
 
-               * - ``GitPython``
-                 - `GitPython <https://github.com/gitpython-developers/GitPython>`_ 3.1.44
+                  * - ``GitPython``
+                    - `GitPython <https://github.com/gitpython-developers/GitPython>`_ 3.1.44
 
-               * - ``opencv-python-headless``
-                 - `opencv-python-headless <https://pypi.org/project/opencv-python-headless/>`_ 4.10.0.84
+                  * - ``opencv-python-headless``
+                    - `opencv-python-headless <https://pypi.org/project/opencv-python-headless/>`_ 4.10.0.84
 
-               * - ``peft``
-                 - `PEFT <https://huggingface.co/docs/peft/en/index>`_ 0.14.0
+                  * - ``peft``
+                    - `PEFT <https://huggingface.co/docs/peft/en/index>`_ 0.14.0
 
-               * - ``protobuf``
-                 - `Protocol Buffers <https://github.com/protocolbuffers/protobuf>`_ 5.29.2
+                  * - ``protobuf``
+                    - `Protocol Buffers <https://github.com/protocolbuffers/protobuf>`_ 5.29.2
 
-               * - ``pytest``
-                 - `PyTest <https://docs.pytest.org/en/stable/>`_ 8.3.4
+                  * - ``pytest``
+                    - `PyTest <https://docs.pytest.org/en/stable/>`_ 8.3.4
 
-               * - ``python-dotenv``
-                 - `python-dotenv <https://pypi.org/project/python-dotenv/>`_ 1.0.1
+                  * - ``python-dotenv``
+                    - `python-dotenv <https://pypi.org/project/python-dotenv/>`_ 1.0.1
 
-               * - ``seaborn``
-                 - `Seaborn <https://seaborn.pydata.org/>`_ 0.13.2
+                  * - ``seaborn``
+                    - `Seaborn <https://seaborn.pydata.org/>`_ 0.13.2
 
-               * - ``transformers``
-                 - `Transformers <https://huggingface.co/docs/transformers/en/index>`_ 4.47.0
+                  * - ``transformers``
+                    - `Transformers <https://huggingface.co/docs/transformers/en/index>`_ 4.47.0
 
-         ``pytorch_benchmark_setup.sh`` downloads the following datasets from Hugging Face:
+            ``pytorch_benchmark_setup.sh`` downloads the following datasets from Hugging Face:
 
-         * `bghira/pseudo-camera-10k <https://huggingface.co/datasets/bghira/pseudo-camera-10k>`_
+            * `bghira/pseudo-camera-10k <https://huggingface.co/datasets/bghira/pseudo-camera-10k>`_
 
    {% for model_group in model_groups %}
       {% for model in model_group.models %}
-         {% if model_group.tag == "pre-training" and model.mad_tag in ["pyt_train_llama-3.1-8b", "pyt_train_llama-3.1-70b", "pyt_train_flux"] %}
+         {% set training_modes = model.training_modes %}
+         {% set training_mode_descs = {
+            "pretrain": "Benchmark pre-training.",
+            "HF_pretrain": "Llama 3.1 8B pre-training with FP8 precision."
+         } %}
+         {% set available_modes = training_modes | select("in", ["pretrain", "HF_pretrain"]) | list %}
+         {% if available_modes %}
 
          .. container:: model-doc {{ model.mad_tag }}
 
-            .. rubric:: Pretraining
+            .. rubric:: Pre-training
 
             To start the pre-training benchmark, use the following command with the
             appropriate options. See the following list of options and their descriptions.
 
             .. code-block:: shell
 
-               ./pytorch_benchmark_report.sh -t pretrain -m {{ model.model_repo }} -p $datatype -s $sequence_length
-
-            .. list-table::
-               :header-rows: 1
-
-               * - Name
-                 - Options
-                 - Description
-
-            {% if model.mad_tag == "pyt_train_llama-3.1-8b" %}
-               * - ``$datatype``
-                 - ``BF16`` or ``FP8``
-                 - Only Llama 3.1 8B supports FP8 precision.
-            {% else %}
-               * - ``$datatype``
-                 - ``BF16``
-                 - Only Llama 3.1 8B supports FP8 precision.
-            {% endif %}
-
-               * - ``$sequence_length``
-                 - Sequence length for the language model.
-                 - Between 2048 and 8192. 8192 by default.
+               ./pytorch_benchmark_report.sh -t {% if available_modes | length == 1 %}{{ available_modes[0] }}{% else %}$training_mode{% endif %} \
+                   -m {{ model.model_repo }} \
+                   -p $datatype \
+                   -s $sequence_length
 
             {% if model.mad_tag == "pyt_train_flux" %}
             .. container:: model-doc {{ model.mad_tag }}
 
                .. note::
 
+                  Currently, FLUX models are not supported out-of-the-box on {{ unified_docker.pull_tag }}.
+                  To use FLUX, refer to the previous version of the ``pytorch-training`` Docker: :doc:`previous-versions/pytorch-training-v25.6`
+
                   Occasionally, downloading the Flux dataset might fail. In the event of this
                   error, manually download it from Hugging Face at
                   `black-forest-labs/FLUX.1-dev <https://huggingface.co/black-forest-labs/FLUX.1-dev>`_
                   and save it to `/workspace/FluxBenchmark`. This ensures that the test script can access
                   the required dataset.
             {% endif %}
-         {% endif %}
-
-         {% if model_group.tag == "fine-tuning" %}
-         .. container:: model-doc {{ model.mad_tag }}
-
-            .. rubric:: Fine-tuning
-
-            To start the fine-tuning benchmark, use the following command with the
-            appropriate options. See the following list of options and their descriptions.
-
-            .. code-block:: shell
-
-               ./pytorch_benchmark_report.sh -t $training_mode -m {{ model.model_repo }} -p BF16 -s $sequence_length
 
             .. list-table::
                :header-rows: 1
@@ -383,53 +404,143 @@ The following models are pre-optimized for performance on the AMD Instinct MI325
                  - Options
                  - Description
 
-               * - ``$training_mode``
-                 - ``finetune_fw``
-                 - Full weight fine-tuning (BF16 supported)
-
-               * -
-                 - ``finetune_lora``
-                 - LoRA fine-tuning (BF16 supported)
-
-               * -
-                 - ``finetune_qlora``
-                 - QLoRA fine-tuning (BF16 supported)
-
-               * -
-                 - ``HF_finetune_lora``
-                 - LoRA fine-tuning with Hugging Face PEFT
+               {% for mode in available_modes %}
+               * - {% if loop.first %}``$training_mode``{% endif %}
+                 - ``{{ mode }}``
+                 - {{ training_mode_descs[mode] }}
+               {% endfor %}
 
                * - ``$datatype``
-                 - ``BF16``
-                 - All models support BF16.
+                 - ``BF16``{% if model.mad_tag == "pyt_train_llama-3.1-8b" %} or ``FP8``{% endif %}
+                 - Only Llama 3.1 8B supports FP8 precision.
+
+               * - ``$sequence_length``
+                 - Sequence length for the language model.
+                 - Between 2048 and 8192. 8192 by default.
+         {% endif %}
+
+         {% set training_mode_descs = {
+            "finetune_fw": "Full weight fine-tuning (BF16 and FP8 supported).",
+            "finetune_lora": "LoRA fine-tuning (BF16 supported).",
+            "finetune_qlora": "QLoRA fine-tuning (BF16 supported).",
+            "HF_finetune_lora": "LoRA fine-tuning with Hugging Face PEFT.",
+         } %}
+         {% set available_modes = training_modes | select("in", ["finetune_fw", "finetune_lora", "finetune_qlora", "HF_finetune_lora"]) | list %}
+         {% if available_modes %}
+         .. container:: model-doc {{ model.mad_tag }}
+
+            .. rubric:: Fine-tuning
+
+            To start the fine-tuning benchmark, use the following command with the
+            appropriate options. See the following list of options and their descriptions.
+            See :ref:`supported training modes <amd-pytorch-training-supported-training-modes>`.
+
+            .. code-block:: shell
+
+               ./pytorch_benchmark_report.sh -t $training_mode \
+                   -m {{ model.model_repo }} \
+                   -p $datatype \
+                   -s $sequence_length
+
+            .. list-table::
+               :header-rows: 1
+
+               * - Name
+                 - Options
+                 - Description
+
+               {% for mode in available_modes %}
+               * - {% if loop.first %}``$training_mode``{% endif %}
+                 - ``{{ mode }}``
+                 - {{ training_mode_descs[mode] }}
+               {% endfor %}
+
+               * - ``$datatype``
+                 - ``BF16``{% if "finetune_fw" in available_modes %} or ``FP8``{% endif %}
+                 - All models support BF16.{% if "finetune_fw" in available_modes %} FP8 is only available for full weight fine-tuning.{% endif %}
 
                * - ``$sequence_length``
                  - Between 2048 and 16384.
                  - Sequence length for the language model.
 
+            {% if model.mad_tag in ["pyt_train_llama3.2-vision-11b", "pyt_train_llama-3.2-vision-90b"] %}
             .. note::
 
-               {{ model.model }} currently supports the following fine-tuning methods:
+               For LoRA and QLoRA support with vision models (Llama 3.2 11B and 90B),
+               use the following torchtune commit for compatibility:
 
-            {% for method in model.training_modes %}
-               * ``{{ method }}``
-            {% endfor %}
-            {% if model.training_modes|length < 4 %}
+               .. code-block:: shell
+
+                  git checkout 48192e23188b1fc524dd6d127725ceb2348e7f0e
+
+            {% elif model.mad_tag in ["pyt_train_llama-2-7b", "pyt_train_llama-2-13b", "pyt_train_llama-2-70b"] %}
+            .. note::
+
+               You might encounter the following error with Llama 2: ``ValueError: seq_len (16384) of
+               input tensor should be smaller than max_seq_len (4096)``.
+               This error indicates that an input sequence is longer than the model's maximum context window.
+
+               Ensure your tokenized input does not exceed the model's ``max_seq_len`` (4096
+               tokens in this case). You can resolve this by truncating the input or splitting
+               it into smaller chunks before passing it to the model.
+
+               Note on reproducibility: The results in this guide are based on
+               commit ``b4c98ac`` from the upstream
+               `<https://github.com/pytorch/torchtune>`__ repository. For the
+               latest updates, you can use the main branch.
 
-               The upstream `torchtune <https://github.com/pytorch/torchtune>`_ repository
-               does not currently provide YAML configuration files for other combinations of
-               model to fine-tuning method
-               However, you can still configure your own YAML files to enable support for
-               fine-tuning methods not listed here by following existing patterns in the
-               ``/workspace/torchtune/recipes/configs`` directory.
             {% endif %}
          {% endif %}
       {% endfor %}
    {% endfor %}
 
-               .. rubric:: Benchmarking examples
+            .. rubric:: Benchmarking examples
 
-               For examples of benchmarking commands, see `<https://github.com/ROCm/MAD/tree/develop/benchmark/pytorch_train#benchmarking-examples>`__.
+            For examples of benchmarking commands, see `<https://github.com/ROCm/MAD/tree/develop/benchmark/pytorch_train#benchmarking-examples>`__.
+
+Multi-node training
+-------------------
+
+Pre-training
+~~~~~~~~~~~~
+
+Multi-node training with torchtitan is supported. The provided SLURM script is pre-configured for Llama 3 70B.
+
+To launch the training job on a SLURM cluster for Llama 3 70B, run the following commands from the MAD repository.
+
+.. code-block:: shell
+
+   # In the MAD repository
+   cd scripts/pytorch_train
+   sbatch run_slurm_train.sh
+
+Fine-tuning
+~~~~~~~~~~~
+
+Multi-node training with torchtune is supported. The provided SLURM script is pre-configured for Llama 3.3 70B.
+
+To launch the training job on a SLURM cluster for Llama 3.3 70B, run the following commands from the MAD repository.
+
+.. code-block:: shell
+
+   huggingface-cli login # Get access to HF Llama model space
+   huggingface-cli download meta-llama/Llama-3.3-70B-Instruct --local-dir ./models/Llama-3.3-70B-Instruct # Download the Llama 3.3 model locally
+   # In the MAD repository
+   cd scripts/pytorch_train
+   sbatch Torchtune_Multinode.sh
+
+.. note::
+
+   Information regarding benchmark setup:
+
+   * By default, Llama 3.3 70B is fine-tuned using ``alpaca_dataset``.
+   * You can adjust the torchtune `YAML configuration file
+     <https://github.com/pytorch/torchtune/blob/main/recipes/configs/llama3_3/70B_full_multinode.yaml>`__
+     if you're using a different model.
+   * The number of nodes and other parameters can be tuned in the SLURM script ``Torchtune_Multinode.sh``.
+   * Set the ``mounting_paths`` inside the SLURM script.
+
+Once the run is finished, you can find the log files in the ``result_torchtune/`` directory.
 
 Further reading
 ===============

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

@@ -21,6 +21,8 @@ In this guide, you'll learn about:
 
 - Training a model
 
+  - :doc:`With Primus (Megatron-LM backend) <benchmark-docker/primus-megatron>`
+
   - :doc:`With Megatron-LM <benchmark-docker/megatron-lm>`
 
   - :doc:`With PyTorch <benchmark-docker/pytorch-training>`

docs/reference/gpu-arch-specs.rst

@@ -285,7 +285,7 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
           - Radeon AI PRO R9700
           - RDNA4
           - gfx1201
-          - 16
+          - 32
           - 64
           - 32 or 64
           - 128

docs/sphinx/_toc.yml.in

@@ -27,6 +27,24 @@ subtrees:
     title: ROCm on Radeon GPUs
   - file: how-to/deep-learning-rocm.md
     title: Deep learning frameworks
+    subtrees:
+    - entries:
+      - file: compatibility/ml-compatibility/pytorch-compatibility.rst
+        title: PyTorch compatibility
+      - file: compatibility/ml-compatibility/tensorflow-compatibility.rst
+        title: TensorFlow compatibility  
+      - file: compatibility/ml-compatibility/jax-compatibility.rst
+        title: JAX compatibility
+      - file: compatibility/ml-compatibility/verl-compatibility.rst
+        title: verl compatibility  
+      - file: compatibility/ml-compatibility/stanford-megatron-lm-compatibility.rst
+        title: Stanford Megatron-LM compatibility
+      - file: compatibility/ml-compatibility/dgl-compatibility.rst
+        title: DGL compatibility  
+      - file: compatibility/ml-compatibility/megablocks-compatibility.rst
+        title: Megablocks compatibility
+      - file: compatibility/ml-compatibility/taichi-compatibility.rst
+        title: Taichi compatibility 
   - file: how-to/build-rocm.rst
     title: Build ROCm from source
 
@@ -44,8 +62,8 @@ subtrees:
         title: Training
         subtrees:
         - entries:
-          - file: how-to/rocm-for-ai/training/benchmark-docker/megatron-lm.rst
-            title: Train a model with Megatron-LM
+          - file: how-to/rocm-for-ai/training/benchmark-docker/primus-megatron.rst
+            title: Train a model with Primus and Megatron-Core
           - file: how-to/rocm-for-ai/training/benchmark-docker/pytorch-training.rst
             title: Train a model with PyTorch
           - file: how-to/rocm-for-ai/training/benchmark-docker/jax-maxtext.rst

docs/sphinx/requirements.in

@@ -1,4 +1,4 @@
-rocm-docs-core==1.20.1
+rocm-docs-core==1.22.0
 sphinx-reredirects
 sphinx-sitemap
 sphinxcontrib.datatemplates==0.11.0

docs/sphinx/requirements.txt

@@ -23,7 +23,7 @@ beautifulsoup4==4.13.4
     # via pydata-sphinx-theme
 breathe==4.36.0
     # via rocm-docs-core
-certifi==2025.4.26
+certifi==2025.7.14
     # via requests
 cffi==1.17.1
     # via
@@ -35,18 +35,16 @@ click==8.2.1
     # via
     #   jupyter-cache
     #   sphinx-external-toc
-comm==0.2.2
+comm==0.2.3
     # via ipykernel
-cryptography==45.0.3
+cryptography==45.0.5
     # via pyjwt
-debugpy==1.8.14
+debugpy==1.8.15
     # via ipykernel
 decorator==5.2.1
     # via ipython
 defusedxml==0.7.1
     # via sphinxcontrib-datatemplates
-deprecated==1.2.18
-    # via pygithub
 docutils==0.21.2
     # via
     #   myst-parser
@@ -62,7 +60,7 @@ fastjsonschema==2.21.1
     #   rocm-docs-core
 gitdb==4.0.12
     # via gitpython
-gitpython==3.1.44
+gitpython==3.1.45
     # via rocm-docs-core
 greenlet==3.2.3
     # via sqlalchemy
@@ -74,7 +72,7 @@ importlib-metadata==8.7.0
     # via
     #   jupyter-cache
     #   myst-nb
-ipykernel==6.29.5
+ipykernel==6.30.0
     # via myst-nb
 ipython==8.37.0
     # via
@@ -86,7 +84,7 @@ jinja2==3.1.6
     # via
     #   myst-parser
     #   sphinx
-jsonschema==4.24.0
+jsonschema==4.25.0
     # via nbformat
 jsonschema-specifications==2025.4.1
     # via jsonschema
@@ -116,7 +114,7 @@ mdit-py-plugins==0.4.2
     # via myst-parser
 mdurl==0.1.2
     # via markdown-it-py
-myst-nb==1.2.0
+myst-nb==1.3.0
     # via rocm-docs-core
 myst-parser==4.0.1
     # via myst-nb
@@ -134,7 +132,6 @@ nest-asyncio==1.6.0
 packaging==25.0
     # via
     #   ipykernel
-    #   pydata-sphinx-theme
     #   sphinx
 parso==0.8.4
     # via jedi
@@ -152,13 +149,13 @@ pure-eval==0.2.3
     # via stack-data
 pycparser==2.22
     # via cffi
-pydata-sphinx-theme==0.15.4
+pydata-sphinx-theme==0.16.1
     # via
     #   rocm-docs-core
     #   sphinx-book-theme
-pygithub==2.6.1
+pygithub==2.7.0
     # via rocm-docs-core
-pygments==2.19.1
+pygments==2.19.2
     # via
     #   accessible-pygments
     #   ipython
@@ -178,7 +175,7 @@ pyyaml==6.0.2
     #   rocm-docs-core
     #   sphinx-external-toc
     #   sphinxcontrib-datatemplates
-pyzmq==26.4.0
+pyzmq==27.0.0
     # via
     #   ipykernel
     #   jupyter-client
@@ -190,9 +187,9 @@ requests==2.32.4
     # via
     #   pygithub
     #   sphinx
-rocm-docs-core==1.20.1
+rocm-docs-core==1.22.0
     # via -r requirements.in
-rpds-py==0.25.1
+rpds-py==0.26.0
     # via
     #   jsonschema
     #   referencing
@@ -220,7 +217,7 @@ sphinx==8.1.3
     #   sphinx-reredirects
     #   sphinxcontrib-datatemplates
     #   sphinxcontrib-runcmd
-sphinx-book-theme==1.1.4
+sphinx-book-theme==1.1.3
     # via rocm-docs-core
 sphinx-copybutton==0.5.2
     # via rocm-docs-core
@@ -252,7 +249,7 @@ sphinxcontrib-runcmd==0.2.0
     # via sphinxcontrib-datatemplates
 sphinxcontrib-serializinghtml==2.0.0
     # via sphinx
-sqlalchemy==2.0.41
+sqlalchemy==2.0.42
     # via jupyter-cache
 stack-data==0.6.3
     # via ipython
@@ -266,7 +263,6 @@ tornado==6.5.1
     #   jupyter-client
 traitlets==5.14.3
     # via
-    #   comm
     #   ipykernel
     #   ipython
     #   jupyter-client
@@ -274,7 +270,7 @@ traitlets==5.14.3
     #   matplotlib-inline
     #   nbclient
     #   nbformat
-typing-extensions==4.14.0
+typing-extensions==4.14.1
     # via
     #   beautifulsoup4
     #   exceptiongroup
@@ -290,7 +286,5 @@ urllib3==2.5.0
     #   requests
 wcwidth==0.2.13
     # via prompt-toolkit
-wrapt==1.17.2
-    # via deprecated
 zipp==3.23.0
     # via importlib-metadata