"git@developer.sourcefind.cn:kecinstone/2024-pra-vllm.git" did not exist on "c07ece5ca490a90b2b19c33ab7da2d21e015d7bd"
- 30 Oct, 2025 2 commits
-
-
Kshitij Lakhani authored
* Fix: Skip determinism tests for bprop for all sm >=100 Signed-off-by:
Kshitij Lakhani <klakhani@nvidia.com> * Add username to TODO Signed-off-by:
pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
-
Phuong Nguyen authored
fix max jobs Signed-off-by:Phuong Nguyen <phuonguyen@nvidia.com>
-
- 29 Oct, 2025 1 commit
-
-
vthumbe1503 authored
* changes working Signed-off-by:
Varun Thumbe <vthumbe@nvidia.com> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * add support for onnx, minor comments Signed-off-by:
Przemyslaw Tredak <ptrendx@gmail.com> Signed-off-by:
-
- 28 Oct, 2025 1 commit
-
-
Phuong Nguyen authored
* jax norm + te quant Signed-off-by:
-
- 27 Oct, 2025 4 commits
-
-
Przemyslaw Tredak authored
Change the pyTorch installation to CUDA 13 in Build All GitHub action to match the version in the JAX container Signed-off-by:Przemek Tredak <ptredak@nvidia.com>
-
Kirthi Shankar Sivamani authored
* Add checks to submodule during setup and automatically checkout Signed-off-by:
Kirthi Shankar Sivamani <ksivamani@nvidia.com> * fix import and formatting Signed-off-by:
-
Kirthi Shankar Sivamani authored
* Remove nvidia-mathdx dep Signed-off-by:
-
Tong Liu authored
dummy wgrad Signed-off-by:
tongliu <tongliu@nvidia.com> Signed-off-by:
Xin Yao <xiny@nvidia.com> Co-authored-by:
-
- 25 Oct, 2025 1 commit
-
-
Charlene Yang authored
* add max_score for fused/unfused F16 non-CP Signed-off-by:
Charlene Yang <8636796+cyanguwa@users.noreply.github.com> * calculate max per head instead of max over all heads Signed-off-by:
-
- 24 Oct, 2025 2 commits
-
-
jberchtold-nvidia authored
fix checks in unoptimized non-rht fp4 quantize kernel Signed-off-by:Jeremy Berchtold <jberchtold@nvidia.com>
-
buptzyb authored
* support cudagraph dw Signed-off-by:
Robin Zhang <robinz@nvidia.com> * fix lint Signed-off-by:
-
- 23 Oct, 2025 4 commits
-
-
Paweł Gadziński authored
* fix perf issue Signed-off-by:
Pawel Gadzinski <pgadzinski@nvidia.com> * fix Signed-off-by:
-
jberchtold-nvidia authored
* Make SR rng state always 2D (num_devices, 4) Signed-off-by:
-
Przemyslaw Tredak authored
* Added sm_120f to the build Signed-off-by:
Copilot <175728472+Copilot@users.noreply.github.com> Signed-off-by:
Oleg Goncharov <64355998+Oleg-Goncharov@users.noreply.github.com> Signed-off-by:
-
Kirthi Shankar Sivamani authored
Include TE core headers in build Signed-off-by:
-
- 22 Oct, 2025 3 commits
-
-
jberchtold-nvidia authored
Defer cublas check on fp8 gemms until lowering Signed-off-by: -
jberchtold-nvidia authored
* [JAX] Support recipe flags for disabling SR, RHT, and 2D quantization Signed-off-by:
-
Evgeny Tsykunov authored
* rename experimental -> custom_recipes Signed-off-by:
Evgeny <etsykunov@nvidia.com> * Decouple python base classes (api) Signed-off-by:
Evgeny Tsykunov <e.tsykunov@gmail.com> * Update tests/pytorch/test_custom_recipe.py Co-authored-by:
-
- 21 Oct, 2025 4 commits
-
-
Kunlun Li authored
* Add post-processing API for FP8 primary weights to support CUDA Graph Signed-off-by:
kunlunl <kunlunl@nvidia.com> * Add post-processing support for plain pytorch tensors Signed-off-by:
Tim Moon <4406448+timmoon10@users.noreply.github.com> --------- Signed-off-by:
-
jberchtold-nvidia authored
HF login in JAX examples Signed-off-by: -
Oleg Goncharov authored
* Removed activations from NVFP4 CPP tests. Removed CMake debugging flags Signed-off-by:
Oleg Goncharov <ogoncharov@nvidia.com> * Better wording Signed-off-by:
-
Zhongbo Zhu authored
* pipeclean, fix nvfp4 padding of 32 alignment Signed-off-by:
Zhongbo Zhu <zhongboz@nvidia.com> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * numerical test passed Signed-off-by:
Tim Moon <tmoon@nvidia.com> * Fix invalid conversion from tensor to int Signed-off-by:
-
- 20 Oct, 2025 2 commits
-
-
Kirthi Shankar Sivamani authored
* Fix CI failures due to deterministic attention Signed-off-by:
-
fzyzcjy authored
* Update permutation.py Signed-off-by:
fzyzcjy <5236035+fzyzcjy@users.noreply.github.com> * Update permutation.py Signed-off-by:
-
- 18 Oct, 2025 1 commit
-
-
Kirthi Shankar Sivamani authored
* Support wheel build for cuda 13 Signed-off-by:
-
- 17 Oct, 2025 6 commits
-
-
Alp Dener authored
Make `CanonicalizeGemmInput()` support non-TN layout FP8 GEMM on Blackwell with column-wise/transposed data (#2233) Modified CanonicalizeGemmInput() logic to pull from column-wise data for FP8 GEMM on Blackwell when row-wise is not available. Signed-off-by:Alp Dener <adener@nvidia.com>
-
Neil Tenenholtz authored
* Fix test of FSDP2 by correcting init logic and applying autocast This fixes multiple issues in the FSDP2 test, namely 1. Previously fp8 init was performed when `args.fp8_init == False`. I have updated the logic to match what I presume was intended by leveraging the nullcontext context manager. 2. `te.fp8_autocast` was previously not called; the recipe was created but was unused. The autocast context manager now wraps the model's computation. Signed-off-by:
Neil Tenenholtz <ntenenz@users.noreply.github.com> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Fix typo Signed-off-by:
-
Haowen Zheng authored
Signed-off-by:
将来 <jianglai.zhw@alibaba-inc.com> Co-authored-by:
-
Tim Geypens authored
Signed-off-by:Tim Geypens <tim.geypens@gmail.com>
-
Kevin Tong authored
* CUDA RHT Signed-off-by:
Kevin Tong <kevin@augmentcode.com> * Fix cuda graphs Signed-off-by:
-
Kshitij Lakhani authored
* Fix imports in test for deprecated jax.experimental.pjit Signed-off-by:
Kshitij Janardan Lakhani <klakhani@login-eos01.eos.clusters.nvidia.com> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci --------- Signed-off-by:
-
- 16 Oct, 2025 3 commits
-
-
Przemek Tredak authored
Signed-off-by: -
xiaoxi-wangfj authored
* [PyTorch] Add record_stream and untyped_storage func op in QuantizedTensor Signed-off-by:
xiaoxi-wangfj <690912414@qq.com> * Update transformer_engine/pytorch/tensor/float8_blockwise_tensor.py Co-authored-by:
-
Selvaraj Anandaraj authored
Signed-off-by:
Selvaraj Anandaraj <selvaraja@login-ptyche02.ptyche.clusters.nvidia.com> Co-authored-by:
-
- 15 Oct, 2025 2 commits
-
-
Santosh Bhavani authored
* Enhance Latest News section with recent TE and FP8 developments - Adds NVFP4 pretraining research paper with PR #2177 reference Signed-off-by:
Santosh Bhavani <santosh.bhavani@live.com> * update nvfp4 reference Signed-off-by:
-
Paul Gibbons authored
* fixes for start_end_list usage in TE debug Signed-off-by:
Paul Gibbons <pgibbons@nvidia.com> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci --------- Signed-off-by:
Paweł Gadziński <62263673+pggPL@users.noreply.github.com>
-
- 14 Oct, 2025 4 commits
-
-
Tim Moon authored
* Require cuDNN 9.14.0+ for fused attention with FP8 current scaling Signed-off-by:
-
Kirthi Shankar Sivamani authored
* Initial API change Signed-off-by:
-
Kshitij Lakhani authored
* Add BRCM support when creating a test mask for fused attn Signed-off-by:
-
Evgeny Tsykunov authored
* Fix update_quantized in ref nvfp4 quantizer Signed-off-by:
-
