- 03 Oct, 2025 1 commit
-
-
Jan Bielak authored
* Update to_string(NVTEScalingMode) to include block scaling Signed-off-by:
Jan Bielak <jbielak@nvidia.com> * Add `nvte_swizzle_block_scaling_to_mxfp8_scaling_factors` Signed-off-by:
Copilot <175728472+Copilot@users.noreply.github.com> Signed-off-by:
Tim Moon <4406448+timmoon10@users.noreply.github.com> * Apply suggestions from code review Signed-off-by:
pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Co-authored-by:
-
- 02 Oct, 2025 1 commit
-
-
Tim Moon authored
* Make sure to set usages for linear op quantizers before forward Signed-off-by:
Tim Moon <tmoon@nvidia.com> * Avoid unsupported case for fused dbias+quantize kernel Hopper does not support dbias + FP8 cast without FP8 transpose. Signed-off-by:
-
- 01 Oct, 2025 1 commit
-
-
Evgeny Tsykunov authored
* Introduce QuantizerBase Signed-off-by:
Evgeny <etsykunov@nvidia.com> * Expose as a first-class API Signed-off-by:
-
- 30 Sep, 2025 2 commits
-
-
vthumbe1503 authored
* Test working as I think it should work 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 Signed-off-by:
Vladimir Cherepanov <vcherepanov@nvidia.com> * Saving... Signed-off-by:
Przemyslaw Tredak <ptrendx@gmail.com> Signed-off-by:
Vladimir Cherepanov <56651474+mk-61@users.noreply.github.com> * Add license Signed-off-by:
Ming Huang <mingh@nvidia.com> * Slightly refactor Signed-off-by:
Phuong Nguyen <phuonguyen@nvidia.com> Signed-off-by:
Jeremy Berchtold <jberchtold@nvidia.com> Co-authored-by:
jberchtold-nvidia <158520091+jberchtold-nvidia@users.noreply.github.com> Signed-off-by:
Kshitij Lakhani <klakhani@nvidia.com> Signed-off-by:
Charlene Yang <8636796+cyanguwa@users.noreply.github.com> * fix remaining CI failures Signed-off-by:
Pawel Gadzinski <pgadzinski@nvidia.com> * fix Signed-off-by:
zhongboz <zhongboz@nvidia.com> Co-authored-by:
Kirthi Shankar Sivamani <ksivamani@nvidia.com> Signed-off-by:
oliver könig <okoenig@nvidia.com> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Update setup.py Signed-off-by:
Xin Yao <xiny@nvidia.com> * Remove exceptions from destructors Signed-off-by:
Robin Zhang <robinz@nvidia.com> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * update Signed-off-by:
Vasudevan Rengasamy <vrengasamy@nvidia.com> Co-authored-by:
Selvaraj Anandaraj <selvaraja@login-ptyche01.ptyche.clusters.nvidia.com> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Fixed typo Signed-off-by:
Selvaraj Anandaraj <selvaraja@login-preos01.a51.clusters.nvidia.com> * Fixed typo Signed-off-by:
Paweł Gadziński <62263673+pggPL@users.noreply.github.com> Signed-off-by:
djns99 <40156487+djns99@users.noreply.github.com> Signed-off-by:
Casper <casperbh.96@gmail.com> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci Signed-off-by:
Hongbin Liu <hongbinl@nvidia.com> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * fix CI Signed-off-by:
tongliu <tongliu@nvidia.com> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci --------- Signed-off-by:
Jonathan Mitchell <jomitchell@login-eos02.eos.clusters.nvidia.com> * assert line change Signed-off-by:
Sudhakar Singh <sudhakars@nvidia.com> * address review comments Signed-off-by:
Yuzhong Wang <yuzhongw@nvidia.com> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Update transformer_engine/pytorch/tensor/_internal/float8_blockwise_tensor_base.py Signed-off-by:
Min Yang <min.yang@shopee.com> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * refactor: refactor multi tensor gemm interface Signed-off-by:
alan yang <89962857+cassiewilliam@users.noreply.github.com> Co-authored-by:
zhujian <zhujian.whu.cs@gmail.com> Fix cuDNN version checks when getting backend and for sm89 kv cache (#2185) * Fix cudnn version checks for kv cache for sm89. Add cudnn version check in preparation for 9.14 when getting backend Signed-off-by:
-
Charlene Yang authored
* debug existing usage Signed-off-by:
-
- 29 Sep, 2025 1 commit
-
-
Kirthi Shankar Sivamani authored
* Add NVFP4 recipe Signed-off-by:
Frank Sun <frsun@nvidia.com> Co-authored-by:
Oleg Goncharov <ogoncharov@nvidia.com> Co-authored-by:
Teddy Do <tdophung@nvidia.com> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Add MathDx dependency to GitHub builds Signed-off-by:
Przemek Tredak <ptredak@nvidia.com> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Fix linter warning Signed-off-by:
-
- 22 Sep, 2025 1 commit
-
-
Charlene Yang authored
* first draft; debug plan failure Signed-off-by:
Chen Cui <chcui@nvidia.com> Signed-off-by:
-
- 18 Sep, 2025 1 commit
-
-
alan yang authored
* feat: add cutlass group gemm support Signed-off-by:
-
- 17 Sep, 2025 2 commits
-
-
Sudhakar Singh authored
* add tutorial files and other local changes Signed-off-by:
-
vthumbe1503 authored
* Test working as I think it should work Signed-off-by:
-
- 16 Sep, 2025 1 commit
-
-
Jan Bielak authored
* Do not use norm fwd + amax fusion if cudnn backend is requested Signed-off-by:
-
- 08 Sep, 2025 1 commit
-
-
vasunvidia authored
* Fused RoPE with combined QKV input. Initial commit for Dropout with 8-bit RNG Fix documentation Initial commit for Fused QKV RoPE WIP Initial tests passing Enable rotary percent and margin Enable CP2, start_positions, interleaved Cleanup test Revert "Fix documentation" This reverts commit 53df10044e7769982bd4af2ae2628e6b7717e715. Revert "Initial commit for Dropout with 8-bit RNG" This reverts commit 301505e24031cbcd679069e1c2cd4d00eedf2dca. Cleanup. Minor cleanup Signed-off-by:
-
- 04 Sep, 2025 1 commit
-
-
Zhongbo Zhu authored
* add noop to comp amax Signed-off-by:
-
- 31 Aug, 2025 1 commit
-
-
vasunvidia authored
* Add dropout kernel with 8-bit RNG Co-authored-by:
-
- 28 Aug, 2025 1 commit
-
-
Zhongbo Zhu authored
[PyTorch][MOE] Tentative Fix For Replacing from_blob with empty for experts receiving zero tokens (#2134) use torch empty for empty shape instead of from_blob Signed-off-by:
-
- 26 Aug, 2025 1 commit
-
-
Xin Yao authored
expose more activation functions Signed-off-by:
-
- 20 Aug, 2025 1 commit
-
-
Jan Bielak authored
* Add rmsnorm_bwd_add Signed-off-by:
-
- 15 Aug, 2025 1 commit
-
-
Jan Bielak authored
* Add `nvte_cublas_gemm_scaled` Signed-off-by:
-
- 14 Aug, 2025 1 commit
-
-
Tim Moon authored
* Unfused impl for dbias-quantize Signed-off-by:
-
- 09 Aug, 2025 1 commit
-
-
Daniel Stokes authored
* fix: Add stream synchronization before destroying MPI communicator (#1979) Signed-off-by:
-
- 07 Aug, 2025 1 commit
-
-
ldl authored
[PyTorch] Fix numeric overflow caused by int-type parameters and return value in the roundup function (#2034) Signed-off-by:
lvdunlin <lvdunlin@xiaomi.com> Co-authored-by:
-
- 06 Aug, 2025 2 commits
-
-
hx authored
[PyTorch] fix input_quantizer usage for save_original_input; fix blockwise FP8 convert_and_update_tensor (#1978) * fix input_quantizer in save_original_input bwd Signed-off-by:
Hongxiao Bai <hongxiaob@nvidia.com> * fix get shape of blockwise tensor with only compact colwise data Signed-off-by:
-
Xin Yao authored
* for loop Signed-off-by:
-
- 04 Aug, 2025 1 commit
-
-
Tim Moon authored
* Add basic kernel for swapping first two tensor dims Signed-off-by:
-
- 31 Jul, 2025 1 commit
-
-
Jan Bielak authored
* Compute amax in normalization kernels as long as the pointer is provided, even if using non quantized output Signed-off-by:
-
- 30 Jul, 2025 1 commit
-
-
Jan Bielak authored
Refactor normalization.cpp to use quantizer logic introduced in #1952 instead of manual quantization Signed-off-by:
-
- 29 Jul, 2025 3 commits
-
-
Jan Bielak authored
* Compute amax in activation kernels when the output pointer is provided, even for non-fp8 outputs Signed-off-by:
-
Tim Moon authored
* remove reciprocal op Signed-off-by:
-
Przemyslaw Tredak authored
Signed-off-by:
-
- 25 Jul, 2025 1 commit
-
-
Kirthi Shankar Sivamani authored
* Remove deprecated device arg Signed-off-by:
-
- 17 Jul, 2025 1 commit
-
-
hx authored
* save original input Signed-off-by:
-
- 14 Jul, 2025 1 commit
-
-
Autumn1998 authored
* fix underterminsic problem in CI Signed-off-by:
-
- 10 Jul, 2025 1 commit
-
-
Autumn1998 authored
* add router fusion Signed-off-by:
-
- 09 Jul, 2025 1 commit
-
-
Zhongbo Zhu authored
* functional passed Signed-off-by:
-
- 26 Jun, 2025 2 commits
-
-
xiaoxi-wangfj authored
* [PyTorch|common] Implement unpadding kernel for FP8 1. Add multi-tensor unpadding kernel 2. Replace split+cat with unpadding kernel in Fp8Padding and Fp8Unpadding 3. Add unpadding with padding unit tests Signed-off-by:
xiaoxi-wangfj <690912414@qq.com> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * add license Signed-off-by:
-
Zhongbo Zhu authored
* finish python ref impl for bulk alloc Signed-off-by:
-
- 13 Jun, 2025 3 commits
-
-
Charlene Yang authored
* add support for head dim > 128 Signed-off-by:
Charlene Yang <charleney@nvidia.com> * remove debugging Signed-off-by:
-
Oleg Goncharov authored
* Added support of FP4 data type Signed-off-by:
Oleg Goncharov <64355998+Oleg-Goncharov@users.noreply.github.com> * Pass correct num bits to create_2D_tensor_map; fixes CI Signed-off-by:
-
Daniel Stokes authored
* Add support for overlapping wgrad NCCL AG with dgrad GEMM Signed-off-by:
dastokes <dastokes@dastokes-dvt-01.nvidia.com>
-
- 06 Jun, 2025 1 commit
-
-
Phuong Nguyen authored
* refactor the multi_stream utils + implement nvte_multi_tensor_quantize in TE/Common * implement GroupedQuantizer and grouped_quantize in jaxx * fix logical_axes_names for transpose tensor in ScaledTensor Signed-off-by:
Hua Huang <huah@nvidia.com> Co-authored-by:
-
