- 16 Jul, 2025 1 commit
-
-
vcherepanov-nv authored
Signed-off-by:Vladimir Cherepanov <vcherepanov@nvidia.com>
-
- 14 Jul, 2025 4 commits
-
-
Tim Moon authored
* Add run-time version checks in cuBLAS GEMM wrapper Signed-off-by:
Tim Moon <tmoon@nvidia.com> * Add run-time version logic for multicast Signed-off-by:
pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
-
Alp Dener authored
* added XLA FFI custom op for TE/common nvte_cublas_gemm Signed-off-by:
Alp Dener <adener@nvidia.com> started GemmPrimitive, abstract done Signed-off-by:
Phuong Nguyen <phuonguyen@nvidia.com> * fix bug on rhs_scale_inv usage Signed-off-by:
-
Autumn1998 authored
* fix underterminsic problem in CI Signed-off-by:
tongliu <tongliu@nvidia.com> * fix bug on mbs>1 Signed-off-by:
Tim Moon <4406448+timmoon10@users.noreply.github.com> Co-authored-by:
Xin Yao <xiny@nvidia.com>
-
hx authored
* optimize permute 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 lint 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:
-
- 11 Jul, 2025 2 commits
-
-
Phuong Nguyen authored
* memset for the mxfp8 scale padding Signed-off-by:
-
Tim Moon authored
Make MXFP8Tensor unpickling function backward compatible Signed-off-by:
-
- 10 Jul, 2025 3 commits
-
-
Autumn1998 authored
* add router fusion Signed-off-by:
-
buptzyb authored
* Reuse cudagraph input and output tensor memory Signed-off-by:
Robin Zhang <robinz@nvidia.com> * Wrap _make_graphed_callables with fp8 Signed-off-by:
Kirthi Shankar Sivamani <ksivamani@nvidia.com>
-
Paweł Gadziński authored
* push Signed-off-by:
Pawel Gadzinski <pgadzinski@nvidia.com> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * fixes Signed-off-by:
Paweł Gadziński <62263673+pggPL@users.noreply.github.com> * Update tests/pytorch/test_sanity.py Co-authored-by:
-
- 09 Jul, 2025 4 commits
-
-
Selvaraj Anandaraj authored
Signed-off-by:
Selvaraj Anandaraj <selvaraja@login-ptyche02.ptyche.clusters.nvidia.com> Co-authored-by:
-
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:
-
Hua Huang authored
* Fix JAX grouped gemm error on CUDA 12.9.1 & later by using 16B alignment for scale ptr Signed-off-by:
Hua Huang <huah@nvidia.com> * Pad MXFP8 scales with 2*-127 instead of NaNs Signed-off-by:
-
Xin Yao authored
* Fix align_size Signed-off-by:
-
- 08 Jul, 2025 2 commits
-
-
Jan Bielak authored
* Change pre_forward to pre_first_forward Signed-off-by:
-
Jan Bielak authored
Add test for `LayerNormMLP` implementation using `te.ops.Sequential` to `test_fusible_ops.py` (#1924) * Add e2e test for LayerNormMLP implemented with te.Sequential Signed-off-by:
-
- 01 Jul, 2025 3 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:
-
Jan Bielak authored
* Replace `is_float8_tensor` with `is_quantized_tensor` Replace free function `is_float8_tensor` with `is_quantized_tensor` in `_common.py` and use it throughout the `ops` codebase to check if a tensor is a (possibly internal) quantized tensor 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:
-
- 28 Jun, 2025 1 commit
-
-
yuzhongw-nvidia authored
* fix: (1) UT ignores MLA; (2) bshd format runtime error. Ban fp8 mla attn + cp due to correctness problem Signed-off-by:
Yuzhong Wang <yuzhongw@nvidia.com> * only disable FP8 CP for MLA Signed-off-by:
Charlene Yang <8636796+cyanguwa@users.noreply.github.com> --------- Signed-off-by:
-
- 26 Jun, 2025 6 commits
-
-
Charlene Yang authored
* skip kv cache for sm89, cudnn < 9.12 Signed-off-by:
-
Jan Bielak authored
* Optimize _common.reshape by removing redundant operations Signed-off-by:
-
Paweł Gadziński authored
* fixed the bug Signed-off-by:
-
Paweł Gadziński authored
* fix Signed-off-by:
-
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:
-
- 18 Jun, 2025 1 commit
-
-
Phuong Nguyen authored
* TensorUsage + FP8 GEMM with all layouts handling on BW Signed-off-by:
-
- 16 Jun, 2025 2 commits
-
-
Li Tao authored
* Fix an issue when mcore uses te fusion ce implementation Signed-off-by:
lit <lit@nvidia.com> * simplify unit test code Signed-off-by:
-
Hua Huang authored
* Support MXFP8 and handle empty matrices Signed-off-by:
-
- 13 Jun, 2025 7 commits
-
-
Kshitij Lakhani authored
* Add support for Fused Attn MLA head_dim_qk != head_dim_v Modify is_fused_attn_kernel_available() to accept different head_dims for qk and v Modify FusedAttnHelper to accept different head_dims for qk and v and modify assert dims checks in parse_qkv_aval() Modify FusedAttnFwdPrimitive and FusedAttnBwdPrimitive to accept different head_dims for qk and v Modify Fused Attn related cpp and csrc extension API calls to accept different head_dims for qk and v Modify DotProductAttention call() to extract head dims separately for qk and v Modify the FusedAttn Tests to accommodate for API changes in FusedAttn API Add test case for head_dim_qk != head_dim_v (failing) Modify the baseline JAX appropriately to reshape the output vector based on v dims and not q dims Signed-off-by:
Kshitij Janardan Lakhani <klakhani@nvidia.com> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Fix context dims in general DPA in test_fused_attn Signed-off-by:
-
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 <ogoncharov@nvidia.com> * Refactoring to BitsNum in progress 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:
-
Tim Moon authored
* Add FP8 current scaling to te.Sequential tests Signed-off-by:
-
Tim Moon authored
* Do not initialize quantized weights with column-wise usage in inference mode Signed-off-by:
-
Jan Bielak authored
* Flatten basic op params during fuser init 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 3 commits
-
-
Phuong Nguyen authored
* fixes for jittable grouped_quantize * fixes for jittable grouped_gemm * fix contracting_dim for wgrad gemm * exclude jitted grouped_gemm from the unit test as it does not work cudaGraph --------- Signed-off-by:
-
Evgeny Tsykunov authored
* Support L2Norm basic op Signed-off-by:
Evgeny <etsykunov@nvidia.com> * Add L2Norm module wrapper Signed-off-by:
-
Phuong Nguyen authored
* Implemented GroupedDense and TestGroupedDense for BF16, FP16, and FP8 * Fix GroupedGemmFFI cuBLAS workspace alignment bug Signed-off-by:
-
