Commit Graph

832 Commits

Author SHA1 Message Date
Peter Park
85778177a1 Update vLLM docker pull tag 20250415 in vllm-benchmark.rst (#4702) 2025-04-30 16:09:30 -04:00
Istvan Kiss
84177354de Pytorch compatibility page update 2025-04-29 14:43:40 +02:00
Peter Park
7458fcb7ab Update JAX MaxText benchmark doc to v25.5 (#4695)
* fix shell cmd formatting

* add previous versions section

* update docker details and add llama 3.3

* update missed docker image tags to 25.5
2025-04-28 17:52:53 -04:00
Peter Park
16d6e59003 fix link to pytorch-training v25.4 doc (#4696) 2025-04-28 17:52:33 -04:00
Peter Park
a66bc1d85e fix link to previous version in vllm-benchmark.rst (#4689) 2025-04-24 17:54:04 -04:00
Peter Park
36b6ffaf7c Add QwQ 32B to vllm-benchmark.rst (#4685)
* Add Qwen2 MoE 2.7B to vllm-benchmark-models.yaml

* Add QwQ-32B-Preview to vllm-benchmark-models.yaml

* add links to performance results

words

* change "performance validation" to "performance testing"

* remove "-Preview" from QwQ-32B

* move qwen2 MoE after qwen2

* add TunableOp section

* fix formatting

* add link to TunableOp doc

* add tunableop note

* fix vllm-benchmark template

* remove cmdline option for --tunableop on

* update docker details

* remove "training"

* remove qwen2
2025-04-24 16:44:34 -04:00
Peter Park
40e4ba3ecc Update vLLM inference benchmark Docker guide (#4653)
* Remove JAIS 13B and 30B

* update Docker details - vLLM 0.8.3

* add previous version

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

* fix link to previous version
2025-04-24 15:59:13 -04:00
Peter Park
1f41ce26be Add note for chai-1 benchmark Docker in pytorch-inference-benchmark.rst (#4684) 2025-04-24 15:48:53 -04:00
Peter Park
c3faa9670b Add PyTorch inference benchmark Docker guide (+ CLIP and Chai-1) (#4654)
* update vLLM links in deploy-your-model.rst

* add pytorch inference benchmark doc

* update toc and vLLM title

* remove previous versions

* update

* wording

* fix link and "applies to"

* add pytorch to wordlist

* add tunableop note to clip

* make tunableop note appear to all models

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

Co-authored-by: Leo Paoletti <164940351+lpaoletti@users.noreply.github.com>

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

Co-authored-by: Leo Paoletti <164940351+lpaoletti@users.noreply.github.com>

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

Co-authored-by: Leo Paoletti <164940351+lpaoletti@users.noreply.github.com>

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

Co-authored-by: Leo Paoletti <164940351+lpaoletti@users.noreply.github.com>

* fix incorrect links

* wording

* fix wrong docker pull tag

---------

Co-authored-by: Leo Paoletti <164940351+lpaoletti@users.noreply.github.com>
2025-04-23 17:35:52 -04:00
Peter Park
b29b3592bd Update ML framework Docker compatibility docs for 6.4.0 (#4667)
* update pytorch-compatibility.rst

* update tensorflow compat

fix

* update jax and jax-community docker versions
2025-04-22 16:16:16 -04:00
Pratik Basyal
fc162d11e0 6.1.5 column added to historical compatibility develop branch (#4648)
* 6.1.5 column added to historical compatibility
2025-04-17 11:55:32 -04:00
Peter Park
9ff3c2c885 Update PyTorch training Docker doc for 25.5 (#4638)
* update pytorch-training to 25.5

* remove llama 2

* Revert "remove llama 2"

This reverts commit dab672fa7bcbd8bff730382c14177df4301a537d.

* add previous version

* fix run cmd

* add link to docker hub

* fix linting issue

* add Llama 3.3 70B

* update
2025-04-15 18:16:22 -04:00
Peter Park
d057d49af1 Fix vllm Dockerfile.rocm path (#4628) 2025-04-15 11:26:54 -04:00
Peter Park
310864e653 fix link to Dockerfile.rocm (#4573) 2025-04-14 10:10:03 -04:00
Pratik Basyal
af18a170bc Blog link update to 6.4.0 release notes #4596
Blog link update to 6.4.0 release notes
2025-04-11 17:48:42 -04:00
Peter Park
656db2bc84 Update KMD versions in compat matrix (#4594)
* update KMD versions in compat matrix

* update historical compat matrix
2025-04-11 16:48:21 -04:00
Parag Bhandari
493585dfbb Merge branch 'develop' of github.com:ROCm/ROCm into develop 2025-04-11 15:15:43 -04:00
Parag Bhandari
e756d99f65 Merge branch 'develop-internal' into develop 2025-04-11 15:15:19 -04:00
Pratik Basyal
686fcece1d PRE GA Day 640 update for resetting link and HPC application list (#367)
* Links reset to point to latest from stg, internal, RTD, and develop

* ROCm for HPC updated

* GA prep changes
2025-04-11 14:12:57 -05:00
pbhandar-amd
131e34f582 Update w6000-v620.md 2025-04-11 15:11:34 -04:00
Parag Bhandari
db3c46fccf Merge branch 'develop-internal' into develop 2025-04-11 14:32:09 -04:00
pbhandar-amd
7d5ea2f2f9 Update versions.md 2025-04-11 13:16:06 -04:00
pbhandar-amd
18abbbda11 Update versions.md 2025-04-11 13:15:53 -04:00
Peter Park
03137e1146 Remove "preview support" for PyT 2.6 (#368)
* remove pytorch 2.6 preview support note

* update pytorch support release note
2025-04-11 09:12:41 -04:00
Peter Park
8a24176528 Update Thrust and CUB versions for 6.4 + fix compatibility table not displaying (#364)
* Update Thrust and CUB versions

* fix whitespace issue causing build error

* fix onnx runtime ver
2025-04-10 13:38:48 -04:00
Pratik Basyal
1e231b4b28 640 RN known issues batch 4 (#365)
* ROCProfiler deprecation notice udpated

* RHEL 9.6 support removed and 9.5 EOS rejected

* Feedback to KV cache highlight added

* Wrong entry of ROCprofiler-SDK removed

* Additional known issues added

* GA Release date updated

* Consolidated changelog sync
2025-04-10 09:05:34 -04:00
Pratik Basyal
c26f470c8a 6.4.0 Known issues update to RN batch 3 (#362)
* ROCProfiler deprecation notice udpated

* RHEL 9.6 support removed and 9.5 EOS rejected

* Feedback to KV cache highlight added

* Wrong entry of ROCprofiler-SDK removed

* ROCm debugger known issues added

* JAX known issues added

* Ordering fixed

* Compute partition known issues added

* TP sizes known issues added

* Highlight and compatibility matrix updated

* ONNX auto-update corrected

* ROCm systems profiler known issues removed

* Title update
2025-04-09 10:14:14 -04:00
Istvan Kiss
13bd184ec3 Add RDNA4 ISA guide 2025-04-08 13:57:32 +02:00
Istvan Kiss
6c7f167650 Fix broken torchserve link 2025-04-07 16:07:31 +02:00
dependabot[bot]
defb276d93 Build(deps): Bump rocm-docs-core from 1.18.1 to 1.18.2 in /docs/sphinx (#4556)
Bumps [rocm-docs-core](https://github.com/ROCm/rocm-docs-core) from 1.18.1 to 1.18.2.
- [Release notes](https://github.com/ROCm/rocm-docs-core/releases)
- [Changelog](https://github.com/ROCm/rocm-docs-core/blob/develop/CHANGELOG.md)
- [Commits](https://github.com/ROCm/rocm-docs-core/compare/v1.18.1...v1.18.2)

---
updated-dependencies:
- dependency-name: rocm-docs-core
  dependency-version: 1.18.2
  dependency-type: direct:production
  update-type: version-update:semver-patch
...

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2025-04-03 17:02:06 -06:00
Peter Park
fdf24a9c40 fix link to CLR license (#4560) 2025-04-03 13:09:59 -04:00
Dominic Widdows
715cce53de Update workload.rst with small export fix (#4425)
Tiny fix that removes the "export" directive. 
` export HIP_FORCE_DEV_KERNARG=1  hipblaslt-bench ...`
leads to 
bash: export: `hipblaslt-bench': not a valid identifier

whereas just starting with HIP_FORCE_DEV_KERNARG=1 passes this env var to the hipblaslt-bench process, which I think is the intention here.
2025-04-03 13:01:26 -04:00
Jeffrey Novotny
c71201b801 Add Radeon PRO W7800 48GB to GPU hardware specs (#356)
* Add Radeon PRO W7800 48GB to GPU hardware specs

* Adjust row order
2025-04-01 16:44:56 -04:00
Peter Park
ea66bf386a Fix more links in documentation (#4551)
* fix vllm engine args link

* remove RDNA subtree in under system optimization in toc

* fix RDNA 2 architecture PDF link

* fix CLR LICENSE.txt link

* fix rocPyDecode license link
2025-04-01 15:56:34 -04:00
Peter Park
ac2c5e72d4 Fix links in documentation 2025-04-01 15:39:20 -04:00
Peter Park
53eb4f6edb Change AMD SMI ver to 25.3.0 from 25.2.0 (#345) 2025-04-01 13:02:27 -04:00
amitkumar-amd
b178a7ca78 Update the TOC (#355)
* remove 1200

* update link on TOC

* Update docs/sphinx/_toc.yml.in

Co-authored-by: Pratik Basyal <pratik.basyal@amd.com>

---------

Co-authored-by: Pratik Basyal <prbasyal@amd.com>
Co-authored-by: Pratik Basyal <pratik.basyal@amd.com>
2025-03-28 15:59:27 -05:00
Peter Park
424e6148bd Add MaxText training Docker doc
Add MaxText training Docker doc
2025-03-28 11:25:06 -04:00
Peter Park
15aca4be9d Fix ML framework compatible versions for 6.4 (#347)
* Fix ML framework compatible versions for 6.4

* add footnote to historical compat matrix
2025-03-28 10:55:36 -04:00
Pratik Basyal
2cef8b50ae Post RC4 version update and ROCprofiler-SDK changelog update in 640 RN (#349)
* ROCProfiler deprecation notice udpated

* RHEL 9.6 support removed and 9.5 EOS rejected

* Feedback to KV cache highlight added

* Wrong entry of ROCprofiler-SDK removed

* Component version update after RC4

* Typos fixed
2025-03-28 09:30:11 -04:00
Pratik Basyal
a0faccba37 AMD GPU Docs System optimization migration changes in ROCm Docs Develop (#4538)
* AMD GPU Docs System optimization migration changes in ROCm Docs (#296)

* System optimization migration changes in ROCm

* Linting issue fixed

* Linking corrected

* Minor change

* Link updated to Instinct.docs.amd.com

* ROCm docs grid updated by removing IOMMU.rst, pcie-atomics, and oversubscription pages

* Files removed and reference fixed

* Reference text updated

* GPU atomics from 6.4.0 removed
2025-03-27 16:38:10 -04:00
jayfurmanek
f418dbac55 TF updates for 6.4 (#343)
Co-authored-by: Peter Park <peter.park@amd.com>
2025-03-26 14:32:02 -04:00
Pratik Basyal
a412344e0b 640 RN Update with new known issues and OS support (#340)
* ROCProfiler deprecation notice udpated

* RHEL 9.6 support removed and 9.5 EOS rejected

* OS support updated

* Documentation highlight updated

* Update on hardware atomics update

* rocPyDecode version updated

* Quick update in Changes to changes

* Command translation fixed

* gfx950 removed from CK changelog

* glibc version updated

* gfx950 removed

* Changelog list updated
2025-03-26 11:12:30 -04:00
Pratik Basyal
544149631a AMD GPU Docs System optimization migration changes in ROCm Docs (#296)
* System optimization migration changes in ROCm

* Linting issue fixed

* Linking corrected

* Minor change

* Link updated to Instinct.docs.amd.com

* ROCm docs grid updated by removing IOMMU.rst, pcie-atomics, and oversubscription pages

* Files removed and reference fixed

* Reference text updated
2025-03-26 10:01:33 -04:00
dependabot[bot]
1385196fab Build(deps): Bump sphinx-reredirects from 0.1.5 to 0.1.6 in /docs/sphinx (#4527)
Bumps [sphinx-reredirects](https://github.com/documatt/sphinx-reredirects) from 0.1.5 to 0.1.6.
- [Commits](https://github.com/documatt/sphinx-reredirects/compare/v0.1.5...v0.1.6)

---
updated-dependencies:
- dependency-name: sphinx-reredirects
  dependency-type: direct:production
  update-type: version-update:semver-patch
...

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2025-03-25 13:21:53 -06:00
Pruthvi Madugundu
6d6da2c5d6 Update for pytorch 2.6 (#339)
* Update for pytorch 2.6

* add footnote for pyt 2.6

---------

Co-authored-by: Peter Park <peter.park@amd.com>
2025-03-25 15:11:57 -04:00
Peter Park
58d42ec50b Improve "tuning guides" landing page (#4504)
* Improve "tuning guides" landing page

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

Co-authored-by: Pratik Basyal <pratik.basyal@amd.com>

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

Co-authored-by: Pratik Basyal <pratik.basyal@amd.com>

* change tuning to optimization

---------

Co-authored-by: Pratik Basyal <pratik.basyal@amd.com>
2025-03-25 13:54:27 -04:00
Pratik Basyal
796868c8cf 640 OS Support and Compatibility update preGA (#336)
* ROCProfiler deprecation notice udpated

* RHEL 9.6 support removed and 9.5 EOS rejected

* KMD note updated

* rocWMMA content added

* RHEL 9.5 removed form offline installer

* Feedback on ROCm compute profiler added

* Removed updated tensile document entry from Tensile changelog

* Minor change
2025-03-25 09:36:35 -04:00
dependabot[bot]
e396b4898f Build(deps): Bump jinja2 from 3.1.5 to 3.1.6 in /docs/sphinx (#4465)
Bumps [jinja2](https://github.com/pallets/jinja) from 3.1.5 to 3.1.6.
- [Release notes](https://github.com/pallets/jinja/releases)
- [Changelog](https://github.com/pallets/jinja/blob/main/CHANGES.rst)
- [Commits](https://github.com/pallets/jinja/compare/3.1.5...3.1.6)

---
updated-dependencies:
- dependency-name: jinja2
  dependency-type: indirect
...

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2025-03-21 17:18:02 -06:00
Peter Park
8f359da39e Update Megatron-LM doc for 25.4 (#4520)
* update megatron-lm doc

* update 'previous versions'

* add missing space

* update docker pull tag

* Update options and docker pull tag

* Add performance measurements link to megatron-lm doc

* fix previous versions table

* words

* Simplify system validation section

* minor fixes

* fix perv versions tbl
2025-03-21 16:49:55 -04:00