- 01 Aug, 2025 1 commit
-
-
Hongbin Liu authored
* disable wgrad accumulation and reduce in backward() And manually launch it in backward_dw() Signed-off-by:
Hongbin Liu <hongbinl@nvidia.com> * format Signed-off-by:
Kirthi Shankar Sivamani <ksivamani@nvidia.com>
-
- 23 Jul, 2025 1 commit
-
-
Charlene Yang authored
* fix current device for cuDNN/cuBLAS handles Signed-off-by:
Charlene Yang <8636796+cyanguwa@users.noreply.github.com> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * add unit test Signed-off-by:
pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
-
- 22 Jul, 2025 1 commit
-
-
Tim Moon authored
* Debug linear layer when saving original input and using debug quantizer Signed-off-by:
Tim Moon <tmoon@nvidia.com> * Workaround bugs with quantizing with only column-wise usage Signed-off-by:
-
- 21 Jul, 2025 1 commit
-
-
Tim Moon authored
Reset FP8 weight workspace if usages are invalid Signed-off-by:
-
- 17 Jul, 2025 1 commit
-
-
hx authored
* save original input Signed-off-by:
Hongxiao Bai <hongxiaob@nvidia.com> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * fix input_quantizer usage in Linear bwd Signed-off-by:
Xin Yao <xiny@nvidia.com> Co-authored-by:
Tim Moon <4406448+timmoon10@users.noreply.github.com>
-
- 16 Jul, 2025 1 commit
-
-
Paweł Gadziński authored
* some initial code Signed-off-by:
Pawel Gadzinski <pgadzinski@nvidia.com> * onnx support Signed-off-by:
Paweł Gadziński <62263673+pggPL@users.noreply.github.com> * fix Signed-off-by:
root <root@prenyx0221.a51.clusters.nvidia.com> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * lint fix Signed-off-by:
Pawel Gadzinski <pgadzinski@gmail.com> * Update build.yml Signed-off-by:
-
- 12 Jul, 2025 1 commit
-
-
Jan Bielak authored
* Fix clearing tensor data in backward removing is_first_op Signed-off-by:
Jan Bielak <jbielak@nvidia.com> * Misc fixes Signed-off-by:
-
- 09 Jul, 2025 1 commit
-
-
Xin Yao authored
* Fix align_size Signed-off-by:
-
- 01 Jul, 2025 2 commits
-
-
Selvaraj Anandaraj authored
* Added MCore fsdp support for TE Signed-off-by:
Selvaraj Anandaraj <selvaraja@cw-dfw-cs-001-login-01.cm.cluster> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Refactored based on new MCore FSDP Signed-off-by:
-
guyueh1 authored
Fix layernorm output shape in LayernormLinear, make it align with input shape Signed-off-by:
Guyue Huang <guyueh@nvidia.com> Co-authored-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:
zhongboz <zhongboz@nvidia.com> * c++ bulk alloc worked, still draft version Signed-off-by:
-
- 13 Jun, 2025 2 commits
-
-
Tim Moon authored
* Do not initialize quantized weights with column-wise usage in inference mode Signed-off-by:
-
Daniel Stokes authored
* Add support for overlapping wgrad NCCL AG with dgrad GEMM Signed-off-by:
djns99 <40156487+djns99@users.noreply.github.com> * Remove unused wait on memcpy API from UB Signed-off-by:
dastokes <dastokes@dastokes-dvt-01.nvidia.com>
-
- 12 Jun, 2025 2 commits
-
-
Paweł Gadziński authored
typo fix Signed-off-by: -
Kirthi Shankar Sivamani authored
Fix for loading old ckpt formats Signed-off-by:
-
- 07 Jun, 2025 1 commit
-
-
Xin Yao authored
only get skip_fp8_weight_update in fp8_graph_capturing Signed-off-by:
-
- 06 Jun, 2025 2 commits
-
-
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:
Phuong Nguyen <phuonguyen@nvidia.com> --------- Signed-off-by:
Hua Huang <huah@nvidia.com> Co-authored-by:
Ming Huang <mingh@nvidia.com>
-
Zhongbo Zhu authored
[PyTorch] FP8 Subchannel Recipe With FP8 Gather And Configurable Scaling Factor Tensor Swizzling (#1707) * functional kernel for columnwise + no-transpose option, still hacky Signed-off-by:
Keith Wyss <kwyss@nvidia.com> * add more unit test with dequantize compact supported Signed-off-by:
-
- 30 May, 2025 1 commit
-
-
Evgeny Tsykunov authored
* Quantizer update Signed-off-by:
Evgeny Tsykunov <etsykunov@etsykunov-mlt.client.nvidia.com> * Update import Signed-off-by:
Evgeny <etsykunov@nvidia.com> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Introduce _update_weight_quantizers and _get_weight_tensors/_get_weight_quantizers Signed-off-by:
Przemyslaw Tredak <ptredak@nvidia.com>
-
- 20 May, 2025 1 commit
-
-
Peter St. John authored
* Use an empty torch tensor to indicate no fp8 information in extra_state Signed-off-by:
Peter St. John <pstjohn@nvidia.com> * Add huggingface from_pretrained / save_pretrained tests Adds integration tests to ensure models containing TransformerLayer objects can be saved and loaded using the from_pretrained and save_pretrained methods. Signed-off-by:
-
- 19 May, 2025 2 commits
-
-
Evgeny Tsykunov authored
* Check tensor-recipe compatibility Signed-off-by:
-
Paweł Gadziński authored
* tests drop Signed-off-by:
-
- 13 May, 2025 1 commit
-
-
Evgeny Tsykunov authored
* Set sequence_parallel before super().__init__() in norm modules Signed-off-by:
-
- 09 May, 2025 1 commit
-
-
Tim Moon authored
* Avoid spurious warning with non-FP8 GroupedLinear Signed-off-by:
-
- 07 May, 2025 2 commits
-
-
Tim Moon authored
* Initial work toward restoring UB support in te.Sequential Signed-off-by:
-
Peter St. John authored
Signed-off-by:
-
- 06 May, 2025 1 commit
-
-
Przemyslaw Tredak authored
* Changes to Linear Signed-off-by:
-
- 03 May, 2025 1 commit
-
-
Xin Yao authored
* Fix autocast deprecation warnings Signed-off-by:
-
- 01 May, 2025 1 commit
-
-
Tim Moon authored
Fix bug when reshaping norm output from LayerNormLinear Signed-off-by:
-
- 30 Apr, 2025 1 commit
-
-
Paweł Gadziński authored
* init Signed-off-by:
-
- 28 Apr, 2025 2 commits
-
-
Kunlun Li authored
* Prevent using fp8 weights + non-fp8 computation Signed-off-by:
kunlunl <kunlunl@nvidia.com> * Using warnings instead of raising an error Signed-off-by:
-
Alp Dener authored
[PyTorch] Fix cuBLAS workspace leak in applications that initialize+destroy Userbuffers more than once (#1715) safeguarded cuBLAS workspace expansion in initialize_ub() to avoid exponential growth across repeat initializations Signed-off-by:
Alp Dener <adener@nvidia.com> Co-authored-by:
-
- 21 Apr, 2025 1 commit
-
-
Sudhakar Singh authored
* rtx5090 arch fix support Signed-off-by:
Sudhakar Singh <sudhakars@nvidia.com> * apprend `nvte` to the function name so that its visible in framework specific dirs Signed-off-by:
-
- 18 Apr, 2025 1 commit
-
-
Hongbin Liu authored
* split wgrad for GroupedLinear Signed-off-by:
-
- 17 Apr, 2025 2 commits
-
-
wdykas authored
* re merge request Signed-off-by:
Peter Dykas <wdykas@nvidia.com> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * add docstring Signed-off-by:
-
Paweł Gadziński authored
* drop Signed-off-by:
-
- 16 Apr, 2025 1 commit
-
-
Paweł Gadziński authored
* add Signed-off-by:
Przemyslaw Tredak <ptrendx@gmail.com> Signed-off-by:
-
- 14 Apr, 2025 2 commits
-
-
Jianbin Chang authored
* Add fp8 weight transpose cache check in backward, and regenerated it if it does not exist Signed-off-by:
jianbinc <shjwudp@gmail.com> * Properly handle fsdp shard model weight input. Signed-off-by:
-
Xin Yao authored
* Enable MXFP8 and Per-Tensor Current Scaling for Grouped Linear Signed-off-by:
-
