- 09 Jul, 2025 1 commit
-
-
Zhongbo Zhu authored
* functional passed Signed-off-by:
zhongboz <zhongboz@nvidia.com> * before zero padding in mxfp8 swizzle, use torch zeros to malloc for now 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:
Xin Yao <xiny@nvidia.com> * Update padding.cu Signed-off-by:
pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Co-authored-by:
-
Zhongbo Zhu authored
* finish python ref impl for bulk alloc Signed-off-by:
Tim Moon <tmoon@nvidia.com> * Update grouped linear module with fused split+quantize func 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:
Charlene Yang <8636796+cyanguwa@users.noreply.github.com> Co-authored-by:
-
Oleg Goncharov authored
* Added support of FP4 data type Signed-off-by:
Oleg Goncharov <ogoncharov@nvidia.com> * Refactoring to BitsNum in progress Signed-off-by:
Kirthi Shankar Sivamani <ksivamani@nvidia.com> * Refactored the code per the review notes. Fixed JAX build error. 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:
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>
-
- 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:
Tim Moon <4406448+timmoon10@users.noreply.github.com>
-
- 05 Jun, 2025 1 commit
-
-
Kirthi Shankar Sivamani authored
* Fix NVTE_FRAMEWORK=all Signed-off-by:
-
- 29 May, 2025 1 commit
-
-
Przemyslaw Tredak authored
* Changed the Tensor allocation strategy Signed-off-by:
Przemek Tredak <ptredak@nvidia.com> * Fixes Signed-off-by:
-
- 14 May, 2025 1 commit
-
-
Kirthi Shankar Sivamani authored
* rm unused swizzle extensions Signed-off-by:
-
- 13 May, 2025 1 commit
-
-
Kirthi Shankar Sivamani authored
* Disallow kwargs for pybind extensions and release GIL if possible Signed-off-by:
-
- 11 May, 2025 1 commit
-
-
Kirthi Shankar Sivamani authored
* First pass refactor Signed-off-by:
-
- 07 May, 2025 1 commit
-
-
Tim Moon authored
* Initial work toward restoring UB support in te.Sequential Signed-off-by:
-
- 05 May, 2025 1 commit
-
-
Kirthi Shankar Sivamani authored
* Move multi tensors kernels from PyTorch extensions to core Signed-off-by:
-
- 30 Apr, 2025 1 commit
-
-
Kirthi Shankar Sivamani authored
* Move FP8 block scaling kernels to core Signed-off-by:
-
- 29 Apr, 2025 2 commits
-
-
kwyss-nvidia authored
* Reapply "Allow NVTEShape to own data." (#1703) This reverts commit 91405eb4 . Signed-off-by:
-
Xiaowei Ren authored
* fix recompilation of out and lse correction in p2p+bshd/sbhd Signed-off-by:
Xiaowei Ren <xren@nvidia.com> * fix recompilation of get_seq_chunk_ids_for_reordering Signed-off-by:
-
- 22 Apr, 2025 2 commits
-
-
Kirthi Shankar Sivamani authored
* Move radix sort to core Signed-off-by:
-
Sudhakar Singh authored
* add support for `sb1d` freqs tensor in Fused RoPE Signed-off-by:
Sudhakar Singh <sudhakars@nvidia.com> * add `start_positions` variable to `apply_rotary_pos_emb` function to make staggered rope application faster Signed-off-by:
-
- 21 Apr, 2025 1 commit
-
-
Sudhakar Singh authored
* rtx5090 arch fix support Signed-off-by:
-
- 19 Apr, 2025 1 commit
-
-
Tim Moon authored
Revert "Allow NVTEShape to own data. (#1674)" This reverts commit e61ce77c . Signed-off-by:
-
- 18 Apr, 2025 1 commit
-
-
Kunlun Li authored
* Add fp8_primary_weights support for blockwise scaling Signed-off-by:
kunlunl <kunlunl@nvidia.com> custom fsdp Signed-off-by:
-
- 17 Apr, 2025 2 commits
-
-
Xin Yao authored
* move swizzle scaling factor to cpp Signed-off-by:
-
kwyss-nvidia authored
* Allow NVTEShape to own data. Signed-off-by:
-
- 14 Apr, 2025 2 commits
-
-
Xin Yao authored
* Enable MXFP8 and Per-Tensor Current Scaling for Grouped Linear Signed-off-by:
-
Autumn1998 authored
* add support for new recipe on permute_fusion, rm fp unpermute Signed-off-by:
tongliu <tongliu@nvidia.com> * fix lint Signed-off-by:
-
- 10 Apr, 2025 1 commit
-
-
kwyss-nvidia authored
* Add GEMM logic for blockwise quantized tensors. GEMM test cases included in pytorch integration. Signed-off-by:
Xin Yao <yaox12@outlook.com> * Update test_float8blockwisetensor.py Signed-off-by:
-
- 07 Apr, 2025 2 commits
-
-
Xin Yao authored
* fix cpp warning Signed-off-by:
-
Xin Yao authored
* refactor to add cp support for sbhd/bshd Signed-off-by:
-
- 04 Apr, 2025 2 commits
-
-
kwyss-nvidia authored
* Blockwise float8 quantizer and quantized tensor class. The classes are configurable for 128x128 blocksize and 1x128 blocksize via setting block_scaling_dim == 2,1 respectively. Scale tensors are stored in a format emenable for matrix multiplication, however the integration of matmul is deferred as a separate story. Fusions of quantization and DBIAS or activation functions are not yet implemented, and the dequantization is currently implemented in torch. Tests for quantization are included in C++ and pytorch layers, with exact comparison to reference quantizer behavior as well as an attempt to hit interesting branches through the API such as tensor creation in pytorch and CPP and dequantization of row and columnwise usage. Two CUDA kernels for quantization are included, and are direct ports of equivalents in the kitchen repository, where a subchannel recipe has been used for end to end training. Signed-off-by:
-
gdengk authored
* add nvshmem based api support Signed-off-by:
gdeng <gdeng@nvidia.com> * fix lint and license issue Signed-off-by:
-
- 31 Mar, 2025 1 commit
-
-
Tim Moon authored
* Handle case where FP8 current scaling quantizer gets default process group Signed-off-by:
-
- 22 Mar, 2025 2 commits
-
-
Kunlun Li authored
* Enable fp8_primary_weights for current scaling Signed-off-by:
Kunlun Li <94586211+kunlunl@users.noreply.github.com> Signed-off-by:
-
Tim Moon authored
* Do not suppress MXFP8 norm in Python wrapper func Signed-off-by:
-
- 18 Mar, 2025 2 commits
-
-
Przemyslaw Tredak authored
* Do not apply bias when apply_bias is False Signed-off-by:
Przemyslaw Tredak <ptrendx@gmail.com> * Fix the Gelu Aux type Signed-off-by:
-
Charlene Yang authored
* add paged attention; test_kv_cache_accuray and test_paged_attn pass Signed-off-by:
-
- 15 Mar, 2025 1 commit
-
-
Li Tao authored
* support tp-comm-overlap in Current Scaling recipe Signed-off-by:
Li Tao <lit@nvidia.com> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * clean Signed-off-by:
-
- 13 Mar, 2025 2 commits
-
-
Zhongbo Zhu authored
* add tex.bgrad_quantize support for CS Signed-off-by:
-
Tim Moon authored
* Delete row-wise data in single-GPU linear forward Signed-off-by:
-
