Commits · a61caafd6c6fe8973db357e60ac04710193d5273 · OpenDAS / TransformerEngine

08 Aug, 2025 2 commits
- Merge branch 'develop_v2.5' of http://10.16.6.30/dcutoolkit/deeplearing/TransformerEngine · a61caafd
  yuguo authored Aug 08, 2025
  
  a61caafd
- [DCU] add NVTE_INT8_SIM_FP8_TENSORWISE · 3eb6ea62
  yuguo authored Aug 08, 2025
  
  3eb6ea62
07 Aug, 2025 2 commits
- Merge branch 'develop_v2.5' of http://10.16.6.30/dcutoolkit/deeplearing/TransformerEngine · af196942
  yuguo authored Aug 07, 2025
  
  af196942
- [DCU] fix channelwise train accumulate bug · 68d6c506
  yuguo authored Aug 07, 2025
  
  68d6c506
06 Aug, 2025 6 commits
- Merge branch 'develop_v2.5' of http://10.16.6.30/dcutoolkit/deeplearing/TransformerEngine · fb6798f1
  yuguo authored Aug 06, 2025
  
  fb6798f1
- [DCU] fix channelwise train oom bug · 4a013bd5
  yuguo authored Aug 06, 2025
  
  4a013bd5
- Merge branch 'develop_v2.5' of http://10.16.6.30/dcutoolkit/deeplearing/TransformerEngine · 468d9d2d
  yuguo authored Aug 06, 2025
  
  468d9d2d
- [DCU] userargs groupedgemm · ddfbdaf4
  yuguo authored Aug 06, 2025
  
  ddfbdaf4
- Merge branch 'develop_v2.5' of http://10.16.6.30/dcutoolkit/deeplearing/TransformerEngine · eadb9886
  yuguo authored Aug 06, 2025
  
  eadb9886
- [DCU] fix batchlinear core dump in 2.5 · eac0d49b
  yuguo authored Aug 06, 2025
  
  eac0d49b
05 Aug, 2025 2 commits
- Merge branch 'develop_v2.5' of http://10.16.6.30/dcutoolkit/deeplearing/TransformerEngine · 6aafc95e
  yuguo authored Aug 05, 2025
  
  6aafc95e
- [DCU] fix all gather usage · d8041744
  yuguo authored Aug 05, 2025
  
  d8041744
31 Jul, 2025 2 commits
- Merge branch 'develop_v2.5' of http://10.16.6.30/dcutoolkit/deeplearing/TransformerEngine · a9e430c0
  yuguo authored Jul 31, 2025
  
  a9e430c0
- [DCU] triton surpport e4m3_nv · a397dcb7
  yuguo authored Jul 31, 2025
  
  a397dcb7
29 Jul, 2025 1 commit
- Merge branch 'main' of http://10.16.6.30/dcutoolkit/deeplearing/TransformerEngine · 9ab6cd98
  yuguo authored Jul 29, 2025
  
  9ab6cd98
25 Jul, 2025 4 commits
- Merge branch 'develop_v2.4' · 84e8ce2f
  wenjh authored Jul 25, 2025
  
  84e8ce2f
- Disable lightop for now · ff1e14af
  wenjh authored Jul 25, 2025
```
Signed-off-by: wenjh <wenjh@sugon.com>
```
  ff1e14af
- Merge branch 'develop_v2.4' · cddbbf59
  wenjh authored Jul 25, 2025
  
  cddbbf59
- TN w8a8 lightop · 4088797f
  wenjh authored Jul 25, 2025
```
Signed-off-by: wenjh <wenjh@sugon.com>
```
  4088797f
22 Jul, 2025 3 commits
- [DCU] fix compile · 782f6092
  yuguo authored Jul 22, 2025
  
  782f6092
- [DCU] fix merge · eac75188
  yuguo authored Jul 22, 2025
  
  eac75188
- Merge commit '7a9a0825' of... · 44740c6c
  yuguo authored Jul 22, 2025
```
Merge commit '7a9a0825' of https://github.com/NVIDIA/TransformerEngine
```
  44740c6c
21 Jul, 2025 1 commit
- Changed VERSION to 2.7.0.dev0 (#1973) · 7a9a0825
  Kshitij Lakhani authored Jul 21, 2025
```
Signed-off-by: Kshitij Janardan Lakhani <klakhani@nvidia.com>
```
  7a9a0825
19 Jul, 2025 1 commit
- [JAX] Update tolerance of distributed layernorm MLP for FP8 (#1971) · ca7407e3
  jberchtold-nvidia authored Jul 18, 2025
```
Update tolerance of distributed layernorm MLP for FP8
Signed-off-by: Jeremy Berchtold <jberchtold@nvidia.com>
```
  ca7407e3
18 Jul, 2025 9 commits

[Test] Enable cuDNN Norm tests in the CPP suite (#1957) · 86c50977

Phuong Nguyen authored Jul 18, 2025



* enable cudnn norm tests
Signed-off-by: Phuong Nguyen <phuonguyen@nvidia.com>

* exclude tests on pre-Hopper
Signed-off-by: Phuong Nguyen <phuonguyen@nvidia.com>

---------
Signed-off-by: Phuong Nguyen <phuonguyen@nvidia.com>

86c50977

[JAX] Set `precision=HIGHEST` for the ref_grouped_gemm impl in the unit test (#1967) · 2d4644b7

Phuong Nguyen authored Jul 18, 2025



* set precision=HIGHEST for the ref_grouped_gemm impl in the unit test
Signed-off-by: Phuong Nguyen <phuonguyen@nvidia.com>


---------
Signed-off-by: Phuong Nguyen <phuonguyen@nvidia.com>

2d4644b7

Merge branch 'develop_v2.4' of http://10.16.6.30/dcutoolkit/deeplearing/TransformerEngine · 8113d9e0
yuguo authored Jul 18, 2025

8113d9e0
[DCU] fix · d9847b6d
yuguo authored Jul 18, 2025

d9847b6d
Merge branch 'develop_v2.4' of http://10.16.6.30/dcutoolkit/deeplearing/TransformerEngine · 93ecbc82
yuguo authored Jul 18, 2025

93ecbc82
[DCU] fix bias gradient · 736e8f8b
yuguo authored Jul 18, 2025

736e8f8b
Merge branch 'develop_v2.4' · bb8cf71b
wenjh authored Jul 18, 2025

bb8cf71b
Optional lightop · f5349823
wenjh authored Jul 18, 2025
```
Signed-off-by: wenjh <wenjh@sugon.com>
```
f5349823

Update cudnn-frontend to 1.13.0 (#1960) · 657c965b

Charlene Yang authored Jul 17, 2025



* update cudnn-frontend to 1.13.0
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* disable 9.11 for a bug
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



* fix selection logic
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



---------
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>

657c965b

17 Jul, 2025 7 commits

[Common] Optimize KV cache related kernels (#1914) · f8933bbf

Charlene Yang authored Jul 17, 2025



* optimize kv_cache reindex and copy kernels
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



* avoid reindexing from python side
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* rename variable from previous commit
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* minor fix
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* minor fix
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

---------
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>

f8933bbf

[JAX] Remove unneccessary MXFP8 scale_inv padding (#1954) · 5350f277

Phuong Nguyen authored Jul 17, 2025



* remove unnecessary padding
Signed-off-by: Phuong Nguyen <phuonguyen@nvidia.com>

* adapt the test_distributed_layernorm byte count
Signed-off-by: Phuong Nguyen <phuonguyen@nvidia.com>


---------
Signed-off-by: Phuong Nguyen <phuonguyen@nvidia.com>

5350f277

[JAX] Tighten Encoder Test tolerances (#1955) · ed75c2b0
Phuong Nguyen authored Jul 17, 2025
```
tighten encoder test tols
Signed-off-by: Phuong Nguyen <phuonguyen@nvidia.com>
```
ed75c2b0
Merge branch 'develop_v2.4' · 429226fb
wenjh authored Jul 17, 2025

429226fb
Use lightop replace w8a8_mutmal_extension · 1036ccfe
wenjh authored Jul 17, 2025
```
Signed-off-by: wenjh <wenjh@sugon.com>
```
1036ccfe
Merge branch 'develop_v2.4' of http://10.16.6.30/dcutoolkit/deeplearing/TransformerEngine · 3b0a1009
yuguo authored Jul 17, 2025

3b0a1009
[DCU] fix · 00738a42
yuguo authored Jul 17, 2025

00738a42