"git@developer.sourcefind.cn:kecinstone/2024-pra-vllm.git" did not exist on "0f90effc660317070b88b860eda76bfb3b596701"
- 15 Aug, 2025 1 commit
-
-
wenjh authored
Signed-off-by:wenjh <wenjh@sugon.com>
-
- 08 Aug, 2025 1 commit
-
-
yuguo authored
-
- 15 Jul, 2025 1 commit
-
-
yuguo authored
-
- 05 Jun, 2025 1 commit
-
-
Phuong Nguyen authored
* fix otype for fp8 gemm Signed-off-by:
Phuong Nguyen <phuonguyen@nvidia.com> --------- Signed-off-by:
-
- 04 Jun, 2025 2 commits
- 23 May, 2025 1 commit
-
-
jberchtold-nvidia authored
* Fix env variable name in test.sh scripts to properly test pure-JAX implementations Signed-off-by:
Jeremy Berchtold <jberchtold@nvidia.com> * Update test scripts to use pure-JAX impl in encoder test_custom_call_compute.py already uses pure-JAX impl as reference so testing the pure-JAX impl against itself would be redundant. The encoder tests have their own implementation so testing the pure-JAX impl of primitives is still useful. Signed-off-by:
jberchtold-nvidia <158520091+jberchtold-nvidia@users.noreply.github.com> --------- Signed-off-by:
-
- 20 May, 2025 2 commits
-
-
Paweł Gadziński authored
* docs drop Signed-off-by:
Pawel Gadzinski <pgadzinski@nvidia.com> * a Signed-off-by:
Przemyslaw Tredak <ptrendx@gmail.com> Signed-off-by:
Paweł Gadziński <62263673+pggPL@users.noreply.github.com> * Update docs/debug/1_getting_started.rst Co-authored-by:
-
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:
Kirthi Shankar Sivamani <ksivamani@nvidia.com>
-
- 19 May, 2025 1 commit
-
-
Paweł Gadziński authored
* tests drop Signed-off-by:
Przemek Tredak <ptredak@nvidia.com> Co-authored-by:
pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Co-authored-by:
-
- 14 May, 2025 2 commits
-
-
Charlene Yang authored
* reduce FA versions to make CI leaner Signed-off-by:
Charlene Yang <8636796+cyanguwa@users.noreply.github.com> * improve build speed Signed-off-by:
-
Tim Moon authored
* Disable verbose debug logs in CI Signed-off-by:
Tim Moon <tmoon@nvidia.com> * Disable log_cli option Signed-off-by:
Tim Moon <4406448+timmoon10@users.noreply.github.com> --------- Signed-off-by:
-
- 08 May, 2025 1 commit
-
-
yuguo authored
-
- 07 May, 2025 1 commit
-
-
Tim Moon authored
* Initial work toward restoring UB support in te.Sequential Signed-off-by:
-
- 06 May, 2025 1 commit
-
-
jberchtold-nvidia authored
* Fix L2 test_custom_call_compute.py L2 tests Signed-off-by:
-
- 05 May, 2025 1 commit
-
-
Kirthi Shankar Sivamani authored
* Move multi tensors kernels from PyTorch extensions to core Signed-off-by:
-
- 18 Apr, 2025 1 commit
-
-
Phuong Nguyen authored
rm pax/praxis Signed-off-by:
-
- 17 Apr, 2025 1 commit
-
-
linxiddd authored
* [QA] Add error handling - Standardize test failure handling using the unified 'test_fail' function and 'error_exit' function Signed-off-by:
Linxi Ding <linxid@nvidia.com> * Add XML log generation for pytest results - Add `--junitxml` option to pytest command to generate JUnit XML format logs Signed-off-by:
-
- 15 Apr, 2025 2 commits
-
-
Paweł Gadziński authored
* test change Signed-off-by:
-
Phuong Nguyen authored
* script improvement * add wait * add return code back * relax tols for FP8 test in test_multiprocessing_ by 0.001 --------- Signed-off-by:
-
- 12 Apr, 2025 1 commit
-
-
linxiddd authored
[QA] Add error handling - Standardize test failure handling using the unified 'test_fail' function and 'error_exit' function Signed-off-by:
-
- 09 Apr, 2025 2 commits
-
-
Phuong Nguyen authored
* scaling enum abstract * rm NVTE_ from ScalingMode names * rework scaling mode enum in grouped gemm * fix norm sharding --------- Signed-off-by: -
yuguo authored
-
- 07 Apr, 2025 1 commit
-
-
kwyss-nvidia authored
* Add GEMM logic for blockwise quantized tensors. GEMM test cases included in pytorch integration. Signed-off-by:
Keith Wyss <kwyss@nvidia.com> * Update NVTE_BLOCK_SCALING for GEMM. Signed-off-by:
-
- 04 Apr, 2025 1 commit
-
-
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:
-
- 01 Apr, 2025 1 commit
-
-
Phuong Nguyen authored
* refactor + mxfp8 * added grouped gemm * rename linear to dense * added cublas init phase for groupedGemm * relax the tol of test encoder multiprocessing mxfp8 by 0.001 Signed-off-by:
Hua Huang <huah@nvidia.com> Co-authored-by:
-
- 25 Mar, 2025 1 commit
-
-
Charlene Yang authored
* skip cuDNN 9.8 for KV caching Signed-off-by:
-
- 22 Mar, 2025 2 commits
-
-
Kunlun Li authored
* Enable fp8_primary_weights for current scaling Signed-off-by:
kunlunl <kunlunl@nvidia.com> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Use different cast_master_weights_to_fp8 functions depending on the type of quantizer 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:
-
- 19 Mar, 2025 1 commit
-
-
Evgeny Tsykunov authored
Parallelize CPU reference implementation in tests Signed-off-by:
Evgeny Tsykunov <etsykunov@etsykunov-mlt.client.nvidia.com> Co-authored-by:
Oleg Goncharov <64355998+Oleg-Goncharov@users.noreply.github.com>
-
- 18 Mar, 2025 1 commit
-
-
Charlene Yang authored
* add paged attention; test_kv_cache_accuray and test_paged_attn pass Signed-off-by:
Charlene Yang <charleney@nvidia.com> * WIP: non-paged Signed-off-by:
-
- 17 Mar, 2025 1 commit
-
-
linxiddd authored
* [QA] Add error handling -Standardize test failure handling using the unified 'test_fail' function and 'error_exit' function. Signed-off-by:
-
- 13 Mar, 2025 1 commit
-
-
Tim Moon authored
* Explicitly use python3 and pip3 Signed-off-by:
-
- 10 Mar, 2025 1 commit
-
-
Tim Moon authored
Remove Megatron-LM convergence test Signed-off-by:
-
- 09 Mar, 2025 1 commit
-
-
Selvaraj Anandaraj authored
* Verified TE2.0 with offloading Signed-off-by:
Selvaraj Anandaraj <selvaraja@login-eos02.eos.clusters.nvidia.com> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Skipping tests for Ampere and removed child class preparing Signed-off-by:
Selvaraj Anandaraj <selvaraja@login-preos01.a51.clusters.nvidia.com> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * offloading support for MXFP8 dtype Signed-off-by:
root <root@prenyx0095.a51.clusters.nvidia.com> * Hopper backward compatibility cleanup Signed-off-by:
Selvaraj Anandaraj <anandaraj@wisc.edu> Signed-off-by:
-
- 08 Mar, 2025 1 commit
-
-
Zhongbo Zhu authored
* check in per-tensor current scaling full recipe Signed-off-by:
zhongboz <zhongboz@nvidia.com> [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci Signed-off-by:
-
- 05 Mar, 2025 1 commit
-
-
Tim Moon authored
Move Lightning-Thunder integration test to L1 Signed-off-by:
-
- 04 Mar, 2025 1 commit
-
-
Kirthi Shankar Sivamani authored
Signed-off-by:
-
- 28 Feb, 2025 1 commit
-
-
Kirthi Shankar Sivamani authored
* Enforce torch 2.0 and run attn tests with torch.compile Signed-off-by:
-
- 26 Feb, 2025 1 commit
-
-
Selvaraj Anandaraj authored
* Added parallel cross entropy loss implementation using online softmax 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 * Added tests Signed-off-by:
-
