Movatterモバイル変換


[0]ホーム

URL:


Skip to content

Navigation Menu

Sign in
Appearance settings

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly

Sign up
Appearance settings

[TRTLLM-4629] [feat] Add support of CUDA13 and sm103 devices#7568

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to ourterms of service andprivacy statement. We’ll occasionally send you account related emails.

Already on GitHub?Sign in to your account

Merged
litaotju merged 108 commits intomainfromfeat/b300_cu13
Sep 16, 2025
Merged
Changes from1 commit
Commits
Show all changes
108 commits
Select commitHold shift + click to select a range
303604f
upgrade to base image and new TRT, fix many dependency issues
VALLIS-NERIAJun 17, 2025
5c09dc8
CUDA13 breaking changes: c++ compile successful
VALLIS-NERIAJun 17, 2025
1b84604
fix kernel select code to recognize sm103/sm100f
VALLIS-NERIAJul 2, 2025
3a94d80
Update SM100f cubins
Tom-ZhengJul 2, 2025
469a38d
feat: Add support for SM103 3xFP4 tile shapes
djns99Jul 8, 2025
52ad443
disable 3xfp4
VALLIS-NERIAJul 21, 2025
345c2bc
update trtllm-gen sm100f cubins of gemm kernels
VALLIS-NERIAAug 4, 2025
e27cbb5
Ampere moe kernel should build to all arch
VALLIS-NERIAAug 4, 2025
78a55b8
fix vicuna dependency
VALLIS-NERIAAug 4, 2025
271916d
fix deep_gemm & CUDA13
VALLIS-NERIAAug 5, 2025
886437d
merge existing env fix
VALLIS-NERIAAug 6, 2025
b782b6e
fix sm check of kv reuse and chunked context
VALLIS-NERIAAug 6, 2025
84f96b4
update triton and fix deepgemm pip
VALLIS-NERIAAug 6, 2025
759e7a0
Merge remote-tracking branch 'gitlab/main' into feat/gb110_bringup
VALLIS-NERIAAug 6, 2025
bee1df9
remove deepgemm war
VALLIS-NERIAAug 6, 2025
97a3788
update triton image
VALLIS-NERIAAug 6, 2025
ebec4ea
infra: upgrade to DLFW 25.08-pre and TRT 10.13.2.4
ZhanruiSunChAug 12, 2025
36f2e88
Merge branch 'user/zhanruis/update_dlfw_and_cu13' into 'feat/b300_cu13'
ZhanruiSunChAug 12, 2025
0bf6a18
Fix and waive to clean L0
VALLIS-NERIAAug 15, 2025
f12a90b
Merge branch 'feat/gb110_bringup' into 'feat/b300_cu13'
VALLIS-NERIAAug 15, 2025
8c99853
infra: Support build for both CU12 and CU13
ZhanruiSunChAug 18, 2025
c1014e8
Merge branch 'user/zhanruis/update_dlfw_and_cu13_2' into 'feat/b300_c…
ZhanruiSunChAug 18, 2025
4a95d88
revert tlg kernels for ease of merge
VALLIS-NERIAAug 19, 2025
8b53236
Merge remote-tracking branch 'gitlab/main' into user/xiweny/merge_mai…
VALLIS-NERIAAug 19, 2025
5391191
update tg cubins (temp ver)
VALLIS-NERIAAug 21, 2025
f4de884
Merge remote-tracking branch 'gitlab/main' into user/xiweny/merge_mai…
VALLIS-NERIAAug 21, 2025
b7cc06c
disable merge waive list stage
VALLIS-NERIAAug 21, 2025
fa8b52e
fix more sm version check
VALLIS-NERIAAug 22, 2025
808059d
Merge remote-tracking branch 'gitlab/main' into user/xiweny/merge_mai…
VALLIS-NERIAAug 23, 2025
90a9bc4
fix build error
VALLIS-NERIAAug 23, 2025
80ea062
fix cubins
VALLIS-NERIAAug 24, 2025
66b1d8d
Update flashinfer
VALLIS-NERIAAug 25, 2025
ab7febd
Merge commit '31979aefacbf80d2742c98ef30385db162788c84' into feat/b30…
VALLIS-NERIAAug 26, 2025
b1c6f6a
update cutlass and DeepGEMM
VALLIS-NERIAAug 27, 2025
9ad68de
Merge branch 'user/xiweny/update_cutlass_4.2' into 'feat/b300_cu13'
VALLIS-NERIAAug 27, 2025
ee37589
infra: update DLFW 25.08 GA, triton 25.08 GA
ZhanruiSunChAug 28, 2025
c2e1ad9
Merge branch 'user/zhanruis/update_dlfw_and_cu13_3' into 'feat/b300_c…
ZhanruiSunChAug 28, 2025
6fd765f
[None][fix] fix trtllm moe backend error when running gptoss on b300
jiagancAug 28, 2025
f14c740
Merge branch 'dev-jiaganc-fix-b300-gptoss-trtllm' into 'feat/b300_cu13'
VALLIS-NERIAAug 28, 2025
3c06303
[TRTLLM-7755][infra] Add DGX_B300 and GB300 tests in CI
yiqingy0Aug 29, 2025
c425c12
Merge branch 'user/yiqingy/add_b300_tests' into 'feat/b300_cu13'
yiqingy0Aug 29, 2025
0fb835d
fix cutlass moe not falling back
VALLIS-NERIAAug 30, 2025
8d5a7ea
[https://nvbugs/5443053][fix] Disable finalize fusion when Lora is used
jiagancSep 1, 2025
3cc2591
Merge branch 'dev-jiaganc-fix-b300-moe-lora' into 'feat/b300_cu13'
VALLIS-NERIASep 1, 2025
3805f61
[https://nvbugs/5453949][infra] unwaive test_llama_eagle3
bo-nvAug 27, 2025
a765ee4
Merge branch 'feat/b300_cu13-latest' into 'feat/b300_cu13'
VALLIS-NERIASep 1, 2025
14154ec
disable sm103 moe kernel
VALLIS-NERIASep 1, 2025
38ef850
Merge remote-tracking branch 'gitlab/main' into user/xiweny/merge_0901
VALLIS-NERIASep 1, 2025
62a7897
Merge remote-tracking branch 'origin/main' into user/xiweny/merge_0901
VALLIS-NERIASep 2, 2025
90ce786
Fix arg name in _test_trtllm_serve_multimodal_benchmark.py
VALLIS-NERIASep 2, 2025
5bd50d4
update mha cubins and support 103a
VALLIS-NERIASep 3, 2025
1978227
Merge branch 'user/xiweny/mha_103' into 'feat/b300_cu13'
VALLIS-NERIASep 3, 2025
5ca3376
Support DLFW sanity check use CU13 image
ZhanruiSunChSep 5, 2025
9ae01a8
Merge branch 'user/zhanruis/0828_support_cuda_13_for_sanity_check' in…
ZhanruiSunChSep 5, 2025
973fd37
add 3xfp4 cutlass gemm
VALLIS-NERIASep 5, 2025
fcf413e
Merge branch 'user/xiweny/3xfp4_gemm' into 'feat/b300_cu13'
VALLIS-NERIASep 5, 2025
5d4f7f4
update flashinfer and waive bug
VALLIS-NERIASep 5, 2025
22219bc
Add B300 & GB300 CI
VALLIS-NERIASep 5, 2025
2c3f4cb
Merge remote-tracking branch 'origin/main' into feat/b300_cu13
VALLIS-NERIASep 5, 2025
f8864b9
update trtllm gemm
VALLIS-NERIASep 5, 2025
cca347e
[TRTLLM-4629] [feat] Step1: trtllm-gen kernels support sm103
VALLIS-NERIASep 5, 2025
5e7aa76
Merge branch 'user/sm103_trtllmgen' into feat/b300_cu13
VALLIS-NERIASep 5, 2025
10af4f4
[TRTLLM-4629] [feat] Step1: trtllm-gen kernels support sm103
VALLIS-NERIASep 5, 2025
1d7979a
fix
VALLIS-NERIASep 5, 2025
3e71ec7
Merge branch 'user/sm103_trtllmgen' into feat/b300_cu13
VALLIS-NERIASep 5, 2025
65f8478
fix trtllm-gen interface change
VALLIS-NERIASep 5, 2025
bec1e71
fix
VALLIS-NERIASep 5, 2025
0b0781f
fix
VALLIS-NERIASep 6, 2025
3d4f49e
fix missing gemm kernels
VALLIS-NERIASep 6, 2025
1150def
Merge branch 'user/sm103_trtllmgen' into feat/b300_cu13
VALLIS-NERIASep 6, 2025
d12eb4b
fix CI build archs
VALLIS-NERIASep 6, 2025
322db71
Merge remote-tracking branch 'origin/main' into feat/b300_cu13
VALLIS-NERIASep 6, 2025
8f8766a
waive
VALLIS-NERIASep 7, 2025
2912908
Merge remote-tracking branch 'origin/main' into feat/b300_cu13
VALLIS-NERIASep 7, 2025
e6bb1fe
remove non-exist cases
VALLIS-NERIASep 7, 2025
77657de
fix build args
VALLIS-NERIASep 8, 2025
d42201e
remove waivers and cleanup
VALLIS-NERIASep 8, 2025
caea58a
increase build memory
VALLIS-NERIASep 8, 2025
d4d9e77
reset build memory
VALLIS-NERIASep 8, 2025
019b1db
fix 5505835
VALLIS-NERIASep 8, 2025
fdaf4e2
Merge remote-tracking branch 'origin/main' into feat/b300_cu13
VALLIS-NERIASep 8, 2025
e30e0c8
waive
VALLIS-NERIASep 8, 2025
4cf9fed
Merge commit 'ed27a72bcf71f7ab0e7137f7999988c9de82386f' into feat/b30…
VALLIS-NERIASep 8, 2025
b573e07
[None][infra] Disable CU12 build to save build time (cost > 5 hours o…
ZhanruiSunChSep 9, 2025
82833fa
address comments
VALLIS-NERIASep 9, 2025
8cc5ea3
add comment
VALLIS-NERIASep 9, 2025
a8b630f
Merge remote-tracking branch 'origin/main' into feat/b300_cu13
VALLIS-NERIASep 9, 2025
2c287d5
don't throw in ctor
VALLIS-NERIASep 9, 2025
11d603b
fix
VALLIS-NERIASep 9, 2025
d16d98c
fix missing change
VALLIS-NERIASep 9, 2025
5f508b7
Merge remote-tracking branch 'origin/main' into feat/b300_cu13
VALLIS-NERIASep 9, 2025
2e61526
fix
VALLIS-NERIASep 10, 2025
0b73a57
refine sm version check
VALLIS-NERIASep 10, 2025
27c73de
add a line of comment
VALLIS-NERIASep 10, 2025
b8d1ee6
exclude sm70
VALLIS-NERIASep 10, 2025
6133354
fix sm check
VALLIS-NERIASep 11, 2025
41d3cf6
Merge remote-tracking branch 'origin/main' into feat/b300_cu13
VALLIS-NERIASep 11, 2025
ced6e74
[None][infra] Remove WAR on feat branch (#7642)
ZhanruiSunChSep 11, 2025
98cbab0
[None][infra] Update images (#7690)
ZhanruiSunChSep 11, 2025
514ebc2
remove sm70 from fmha_v2 completely
VALLIS-NERIASep 12, 2025
9bd8df7
Merge remote-tracking branch 'origin/main' into feat/b300_cu13
VALLIS-NERIASep 12, 2025
ad20048
remove sm72 & 75
VALLIS-NERIASep 14, 2025
93195ec
waive
VALLIS-NERIASep 15, 2025
98d42f9
Merge remote-tracking branch 'origin/main' into feat/b300_cu13
VALLIS-NERIASep 15, 2025
cf74f40
fix testdb
VALLIS-NERIASep 15, 2025
d48e82a
fix testdb
VALLIS-NERIASep 15, 2025
7657d83
fix
VALLIS-NERIASep 15, 2025
0192299
Merge remote-tracking branch 'origin/main' into feat/b300_cu13
VALLIS-NERIASep 16, 2025
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
PrevPrevious commit
NextNext commit
fix
Signed-off-by: Xiwen Yu <13230610+VALLIS-NERIA@users.noreply.github.com>
  • Loading branch information
@VALLIS-NERIA
VALLIS-NERIA committedSep 15, 2025
commit7657d83553ad27ed6e686d6d5d94889b51b8f71e
2 changes: 0 additions & 2 deletionstests/unittest/llmapi/test_llm.py
View file
Open in desktop
Original file line numberDiff line numberDiff line change
Expand Up@@ -1882,7 +1882,6 @@ async def main():
@pytest.mark.parametrize(
"prompt_logprobs, logprobs, return_context_logits, return_generation_logits",
[(2, None, True, False), (None, 2, False, False)])
@pytest.skip(reason="https://nvbugspro.nvidia.com/bug/5516849")
def test_llm_return_logprobs(prompt_logprobs: Optional[int],
logprobs: Optional[int],
return_context_logits: bool,
Expand All@@ -1894,7 +1893,6 @@ def test_llm_return_logprobs(prompt_logprobs: Optional[int],

@pytest.mark.skip(reason="https://nvbugs/5516660")
@force_ampere
@pytest.skip(reason="https://nvbugspro.nvidia.com/bug/5516849")
def test_llm_return_logprobs_streaming():
llm_return_logprobs_test_harness(2, 2, False, True, streaming=True)

Expand Down

[8]ページ先頭

©2009-2025 Movatter.jp