github/ROCm
1
1
Fork 0
mirror of https://github.com/ROCm/ROCm.git synced 2026-02-11 23:15:06 -05:00
Code Issues Packages Projects Releases Wiki Activity
3,336 Commits 322 Branches 75 Tags
bde0b3e9b45847fb4967f256bfd9e5fccca36b37
Commit Graph

226 Commits

Author SHA1 Message Date
randyh62
45bd726f55 Use intersphinx links for deep learning (#5859) 
* Use intersphinx links for deep learning

* Update deep-learning-rocm.rst

remove Taichi

* Update deep-learning-rocm.rst

Change Install link to "link"

* Apply suggestion from @randyh62

OK
 2026-01-16 13:17:47 -08:00
peterjunpark
a745e45dcb Doc update for vLLM refactor #5855 2026-01-15 11:21:38 -05:00
peterjunpark
2dc22ca890 fix(primus-pytorch.rst): FP8 config instead of BF16 (#5839) 2026-01-07 13:49:31 -05:00
peterjunpark
172b0f7c08 Fix inconsistency in xDiT doc 
Fix inconsistency in xDiT doc
 2025-12-29 10:26:25 -05:00
peterjunpark
c67fac78bd Update docs for xDiT diffusion inference 25.13 Docker release (#5820) 
* archive previous version

* add xdit 25.13

* update history index

* add perf results section
 2025-12-29 08:44:45 -05:00
peterjunpark
e0b8ec4dfb Update training docs for Primus/25.11 (#5819) 
* update conf and toc.yml.in

* archive previous versions

archive data files

update anchors

* primus pytorch: remove training batch size args

* update primus megatron run cmds

multi-node

* update primus pytorch

update

* update

update

* update docker tag
 2025-12-29 08:05:47 -05:00
peterjunpark
3a43bacdda Update xdit diffusion inference history (#5808) 
* Update xdit diffusion inference history

* fix
 2025-12-22 11:05:32 -05:00
peterjunpark
48d8fe139b fix link to ROCm PyT docker image (#5803) 2025-12-19 15:47:55 -05:00
peterjunpark
7455fe57b8 clean up formatting in FA2 page (#5795) 2025-12-19 09:21:41 -05:00
peterjunpark
52c0a47e84 Update Flash Attention guidance in "Model acceleration libraries" (#5793) 
* flash attention update

Signed-off-by: seungrok.jung <seungrok.jung@amd.com>

flash attention update

Signed-off-by: seungrok.jung <seungrok.jung@amd.com>

flash attention update

Signed-off-by: seungrok.jung <seungrok.jung@amd.com>

sentence-case heading

* Update docs/how-to/rocm-for-ai/inference-optimization/model-acceleration-libraries.rst

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

* Apply suggestions from code review

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

---------

Co-authored-by: seungrok.jung <seungrok.jung@amd.com>
Co-authored-by: Leo Paoletti <164940351+lpaoletti@users.noreply.github.com>
 2025-12-19 08:48:52 -05:00
peterjunpark
cbab9a465d Update documentation for JAX training MaxText 25.11 release (#5789) 2025-12-18 11:23:58 -05:00
peterjunpark
459283da3c xDiT diffusion inference v25.12 documentation update (#5786) 
* Add xdit-diffusion ROCm docs page.

* Update template formatting and fix sphinx warnings

* Add System Validation section.

* Add sw component versions/commits.

* Update to use latest v25.10 image instead of v25.9

* Update commands and add FLUX instructions.

* Update Flux instructions. Change image tag. Describe as diffusion inference instead of specifically video.

* git rm xdit-video-diffusion.rst

* Docs for v25.12

* Add hyperlinks to components

* Command fixes

* -Diffusers suffix

* Simplify yaml file and cleanup main rst page.

* Spelling, added 'js'

* fix merge conflict

fix

---------

Co-authored-by: Kristoffer <kristoffer.torp@amd.com>
 2025-12-17 10:20:10 -05:00
peterjunpark
1b4f25733d vLLM inference benchmark 1210 (#5776) 
* Archive previous ver

fix anchors

* Update vllm.rst and data yaml for 20251210
 2025-12-17 09:21:57 -05:00
Pratik Basyal
78e8baf147 Taichi removed from ROCm docs [Develop] (#5779) 
* Taichi removed from ROCm docs

* Warnings fixed
 2025-12-16 13:12:40 -05:00
Matt Williams
65a936023b Fixing link redirects (#5758) 
* Update multi-gpu-fine-tuning-and-inference.rst

* Update pytorch-training-v25.6.rst

* Update pytorch-compatibility.rst
 2025-12-10 11:17:59 -05:00
peterjunpark
bf74351e5a Fix Primus PyTorch doc: training.batch_size -> training.local_batch_size (#5748) 2025-12-08 13:35:22 -05:00
yugang-amd
f2067767e0 xdit-diffusion v25.11 docs (#5744) 2025-12-05 17:09:48 -05:00
peterjunpark
453751a86f fix docker hub links for primus:v25.10 (#5738) 2025-12-04 09:17:33 -05:00
peterjunpark
fb644412d5 Update training Docker docs for Primus 25.10 (#5737) 2025-12-04 09:08:00 -05:00
Alex Xu
007f24fe7b Merge remote-tracking branch 'external/develop' into sync-develop-from-external 2025-11-26 10:09:04 -05:00
Pratik Basyal
fb098b6354 Initial changes for 7.1.1 release notes (#622) 
* Changelog and tables updates for 7.1.1 release notes

* Changelog synced

* Naming udpated

* Added upcoming changes for composable kernel

* Update RELEASE.md

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

* Update RELEASE.md

* Highlights udpated for DGL, ROCm-DS, and HIP documentation

* Changelog synced"

* Offline, runfile and ROCm Bandwidth test updated

* CK/AITER highlight added

* Changelog synced

* AI model highlight updated

* PLDM version added

* Changelog updated

* Leo's feedback incorporated

* Compatibility and PLDM versions udpated

* New docs update added

* ROCm resolved issue added

* Review feedback added

* Link added

* PLDM updated

* PLDM table udpated

* Changes

---------

Co-authored-by: spolifroni-amd <Sandra.Polifroni@amd.com>
 2025-11-17 12:09:59 -05:00
peterjunpark
eb956cfc5c Fixed wording related to VLLM_V1_USE_PREFILL_DECODE_ATTENTION (#5605) 
Co-authored-by: Hongxia Yang <hongxia.yang@amd.com>
 2025-11-11 09:22:11 -05:00
peterjunpark
e05cdca54f Fix references to vLLM docs (#5651) 2025-11-11 09:00:07 -05:00
anisha-amd
04c7374f41 Docs: frameworks 25.10 - compatibility - DGL and llama.cpp (#5648) 2025-11-10 15:26:54 -05:00
yugang-amd
674dc355e4 vLLM 10/24 release (#5626) 
* vLLM 10/24 release

* updates per SME inputs

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

Co-authored-by: Jeffrey Novotny <jnovotny@amd.com>

---------

Co-authored-by: Jeffrey Novotny <jnovotny@amd.com>
 2025-11-05 11:13:50 -05:00
peterjunpark
1515fb3779 Revert "Add xdit diffusion docs (#5576)" (#5580) 
This reverts commit 4132a2609c.
 2025-10-27 16:22:28 -04:00
Kristoffer
4132a2609c Add xdit diffusion docs (#5576) 
* Add xdit video diffusion base page.

* Update supported accelerators.

* Remove dependency on mad-tags.

* Update docker pull section.

* Update container launch instructions.

* Improve launch instruction options and layout.

* Add benchmark result outputs.

* Fix wrong HunyuanVideo path

* Finalize instructions.

* Consistent title.

* Make page and side-bar titles the same.

* Updated wordlist. Removed note container reg HF.

* Remove fp8_gemms in command and add release notes.

* Update accelerators naming.

* Add note regarding OOB performance.

* Fix admonition box.

* Overall fixes.
 2025-10-27 14:56:55 +01:00
peterjunpark
35ca027aa4 Fix broken links under rocm-for-ai/ (#5564) 2025-10-23 14:39:58 -04:00
peterjunpark
90c1d9068f add xref to vllm v1 optimization guide in workload.rst (#5560) 2025-10-22 13:47:46 -04:00
peterjunpark
cb8d21a0df Updates to the vLLM optimization guide for MI300X/MI355X (#5554) 
* Expand vLLM optimization guide for MI300X/MI355X with comprehensive AITER coverage. attention backend selection, environment variables (HIP/RCCL/Quick Reduce), parallelism strategies, quantization (FP8/FP4), engine tuning, CUDA graph modes, and multi-node scaling.

Co-authored-by: PinSiang <pinsiang.tan@embeddedllm.com>
Co-authored-by: Hongxia Yang <62075498+hongxiayang@users.noreply.github.com>
Co-authored-by: pinsiangamd <pinsiang.tan@amd.com>
Co-authored-by: Jeffrey Novotny <jnovotny@amd.com>
 2025-10-22 12:54:25 -04:00
peterjunpark
a613bd6824 JAX Maxtext v25.9 doc update (#5532) 
* archive previous version (25.7)

* update docker components list for 25.9

* update template

* update docker pull tag

* update

* fix intro
 2025-10-17 11:31:06 -04:00
peterjunpark
14bb59fca9 Update Megatron/PyTorch Primus 25.9 docs (#5528) 
* add previous versions

* Fix heading levels in pages using embedded templates (#5468)

* update primus-megatron doc

update megatron-lm doc

update templates

fix tab

update primus-megatron model configs

Update primus-pytorch model configs

fix css class

add posttrain to pytorch-training template

update data sheets

update

update

update

update docker tags

* Add known issue and update Primus/Turbo versions

* add primus ver to histories

* update primus ver to 0.1.1

* fix leftovers from merge conflict
 2025-10-16 12:51:30 -04:00
anisha-amd
a98236a4e3 Main Docs: references of accelerator removal and change to GPU (#5495) 
* Docs: references of accelerator removal and change to GPU

Co-authored-by: Leo Paoletti <164940351+lpaoletti@users.noreply.github.com>
Co-authored-by: Pratik Basyal <pratik.basyal@amd.com>
 2025-10-16 11:22:10 -04:00
Pratik Basyal
036aaa2e78 ROCm for HPC topic updated Develop (#5504) 
* ROCm for HPC topic updated

* ROCm for HPC topic udpated

* Minor editorial
 2025-10-10 22:31:51 -04:00
peterjunpark
68e8453ca5 Update vLLM doc for 10/6 release and bump rocm-docs-core to 1.26.0 (#5481) 
* archive previous doc version

* update model/docker data and doc templates

* Update "Reproducing the Docker image"

* fix: truncated commit hash doesn't work for some reason

* bump rocm-docs-core to 1.26.0

* fix numbering

fix

* update docker tag

* update .wordlist.txt
 2025-10-08 16:23:40 -04:00
peterjunpark
eeea0d2180 Fix heading levels in pages using embedded templates (#5468) 2025-10-03 13:33:14 -04:00
anisha-amd
93c6d17922 Docs: frameworks 25.09 - compatibility - FlashInfer and llama.cpp (#5462) 2025-10-02 13:51:36 -04:00
peterjunpark
2e1b4dd5ee Add multi-node setup instructions for training perf Dockers (#5449) 
---------

Co-authored-by: Jeffrey Novotny <jnovotny@amd.com>
 2025-09-30 14:53:38 -04:00
Peter Park
fd59b5fbac fix links in docs (#5446) 2025-09-29 15:27:32 -04:00
Pratik Basyal
d92d9268dc Use of Radeon and Ryzen reference updated [Develop] (#5432) 
* Use of Radeon and Ryzen reference updated

* Pytorch link update
 2025-09-24 19:07:41 -05:00
Peter Park
442d7e4750 Add env var note to vllm.rst for MoE models and fix links in docs (#5415) 
* docs(vllm.rst): add performance note for MoE models

* docs: fix links

update vllm readme link 20250521

fix links
 2025-09-22 15:58:43 -04:00
Peter Park
d92e5b6c12 Update Primus Megatron doc v25.8 (#5396) 
* megatron: update previous versions list

update

wording

* megatron: update rst and yaml

update primus repo link

update mig guide

* update headings and anchors

* megatron: update doc

* update docker hub urls
 2025-09-19 08:09:21 -04:00
Peter Park
9827ba7ff2 docs: MaxText v25.7 patch update (#5372) 
* remove jax 0.6.0 nanoo fp8 caveat note

* reorder maxtext docker images in data sheet
 2025-09-17 16:25:46 -04:00
Peter Park
e8d104124f Fix PyTorch training benchmark doc template (#5357) 
* fix template

* update wordlist
 2025-09-16 17:21:57 -04:00
Peter Park
26f708da87 Add Stable Diffusion XL to PyT training benchmark doc and fix paths in SGLang Disagg Inference doc (#5282) 
* add sdxl to pytorch-training

* fix sphinx warnings

fix links

* fix paths in cmds and links in sglang disagg

* fix col width

* update release highlights

* fix

quickfix
 2025-09-16 16:49:33 -04:00
Peter Park
bab853a0d3 Add NCF to pytorch training benchmark doc (#5352) 
* add previous version (25.6)

* fix template

* Formatting and wording fixes

* add caveats

* update yaml

* add note to pytorch-training

* fix template

* make model name shorter
 2025-09-16 13:29:28 -04:00
Peter Park
d5101532f7 docs: Add SGLang disaggregated P/D inference w/ Mooncake guide (#5335) 
* add main content

* Update content and format

add clarification

update

update data

* fix

fix

fix

* fix: deepseek v3

* add ki

* Update docs/how-to/rocm-for-ai/inference/benchmark-docker/sglang-distributed.rst

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

* Update docs/how-to/rocm-for-ai/inference/benchmark-docker/sglang-distributed.rst

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

* Update docs/how-to/rocm-for-ai/inference/benchmark-docker/sglang-distributed.rst

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

* Update docs/how-to/rocm-for-ai/inference/benchmark-docker/sglang-distributed.rst

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

* Update docs/how-to/rocm-for-ai/inference/benchmark-docker/sglang-distributed.rst

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

* Update docs/how-to/rocm-for-ai/inference/benchmark-docker/sglang-distributed.rst

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

* Update docs/how-to/rocm-for-ai/inference/benchmark-docker/sglang-distributed.rst

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

* Update docs/how-to/rocm-for-ai/inference/benchmark-docker/sglang-distributed.rst

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

* Update docs/how-to/rocm-for-ai/inference/benchmark-docker/sglang-distributed.rst

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

* Update docs/how-to/rocm-for-ai/inference/benchmark-docker/sglang-distributed.rst

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

---------

Co-authored-by: Leo Paoletti <164940351+lpaoletti@users.noreply.github.com>
 2025-09-16 10:33:58 -05:00
Peter Park
ef4e7ca1fe docs(PyTorch training v25.8): Add Primus and update PyTorch training benchmark docs (#5331) 
* pyt: update previous versions list

update conf.py

* pyt: update yaml and rst

update

update toc

* update headings and anchors

* pyt: update doc

* update docker hub urls
 2025-09-16 10:33:53 -05:00
Peter Park
76cb264f34 Update vllm-history.rst with missing 0909 entry (#5308) 2025-09-16 06:54:34 -04:00
Peter Park
7098bdc03b Update vLLM inference benchmark doc for 0909 release (and Sphinx fixes) (#5289) 2025-09-11 15:01:17 -04:00