kanshan/TensorRT-LLMs
1
0
Fork 0
mirror of https://github.com/NVIDIA/TensorRT-LLM.git synced 2026-01-14 06:27:45 +08:00
Code Issues Actions 1 Packages Projects Releases Wiki Activity
1,484 Commits 25 Branches 59 Tags 2.1 GiB
9106b5d9a5
Commit Graph

1484 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
brb-nv
9106b5d9a5
fix: Skip rope scaling for local layers in Gemma3 VLM (#5773) 
Signed-off-by: Balaram Buddharaju <169953907+brb-nv@users.noreply.github.com>
 2025-07-07 13:36:23 +08:00
ruodil
6103466de2
test: fix some test failure and add llama_nemotron models in perf sanity test, add more torch cases (#5693) 
Signed-off-by: ruodil <200874449+ruodil@users.noreply.github.com>
Co-authored-by: Larry <197874197+LarryXFly@users.noreply.github.com>
 2025-07-07 13:11:41 +10:00
Yanchao Lu
aa4d0f04eb
[Infra] - Always use x86 image for the Jenkins agent (#5756) 
Signed-off-by: Yanchao Lu <yanchaol@nvidia.com>
 2025-07-06 10:25:20 +08:00
Iman Tabrizian
518915b5c6
[nvbug/5337601][fix] Fix disagg + speculative decoding (#5558) 
Signed-off-by: Mike Iovine <6158008+mikeiovine@users.noreply.github.com>
Signed-off-by: Iman Tabrizian <10105175+tabrizian@users.noreply.github.com>
Co-authored-by: Mike Iovine <6158008+mikeiovine@users.noreply.github.com>
 2025-07-04 12:52:35 -04:00
Yi Zhang
5ac92bb8ff
[nvbugs/5336321][fix] Enable attention dp = False test case, Fix TRTLLM Gen Moe workspace allocation (#5463) 
Signed-off-by: Yi Zhang <187001205+yizhang-nv@users.noreply.github.com>
Signed-off-by: yizhan <187001205+yizhang-nv@users.noreply.github.com>
 2025-07-04 23:23:41 +09:00
Yiqing Yan
3e44db11c9
[Infra][nvbugs/5370968] - Unwaive l0 test (#5750) 
Signed-off-by: Yiqing Yan <yiqingy@nvidia.com>
 2025-07-04 15:27:53 +08:00
Yukun He
b0354ef43c
[5321981] fix: Fix the Llama3.1 405B hanging issue. (#5698) 
Correct the output shape of the fusedLayerNormPlugin.

Signed-off-by: Yukun He <23156053+hyukn@users.noreply.github.com>
 2025-07-04 12:29:19 +08:00
Yi Zhang
53394e0030
test: Move some of the test from post merge to pre-merge, update dgx b200 test case (#5640) 
Signed-off-by: Yi Zhang <187001205+yizhang-nv@users.noreply.github.com>
 2025-07-04 13:26:53 +09:00
brb-nv
2b66fe8fbd
[nvbug/5341178][fix] Fix OOM in Llama 4 accuracy test (#5735) 
Signed-off-by: Balaram Buddharaju <169953907+brb-nv@users.noreply.github.com>
 2025-07-04 10:55:34 +08:00
Dom Brown
2aacdba1e4
[TRTLLM-6100] fix: Nvbug 5356427: autotuned TRTLLM Gen fp8 block scale MoE illegal memory access (#5676) 
Signed-off-by: Dom Brown <3886319+DomBrown@users.noreply.github.com>
 2025-07-04 10:38:08 +08:00
Faraz
8a8d2e9901
[NVBUG:5355009] Modify check for fuse_fp4_quant on SM120 (#5651) 
Signed-off-by: Faraz Khoubsirat <58580514+farazkh80@users.noreply.github.com>
Signed-off-by: peaceh <103117813+peaceh-nv@users.noreply.github.com>
Co-authored-by: peaceh-nv <103117813+peaceh-nv@users.noreply.github.com>
 2025-07-03 22:08:15 +09:00
Linda
14f938e510
Doc: Update invalid hugging face URLs (#5683) 
Signed-off-by: Linda-Stadter <57756729+Linda-Stadter@users.noreply.github.com>
 2025-07-03 09:37:01 +02:00
Emma Qiao
2f9d0619c3
[Infra] - Waive failed cases on release/0.21 (#5674) 
Signed-off-by: qqiao <qqiao@nvidia.com>
 2025-07-02 22:23:54 -04:00
brb-nv
a3c0cf02ce
fix: Investigate Gemma3 1B decoder output discrepancy (#5564) 
Signed-off-by: Balaram Buddharaju <169953907+brb-nv@users.noreply.github.com>
 2025-07-03 09:55:25 +08:00
Frank
92d3a2d0e0
[https://nvbugspro.nvidia.com/bug/5351333][fix] Update to chunking calculation. (#5625) 
Signed-off-by: Frank Di Natale <3429989+FrankD412@users.noreply.github.com>
 2025-07-02 17:48:02 +08:00
bhsueh_NV
d5606b062a
fix: [https://nvbugs/5355219] Fix bug of Qwen3 235B CI on dgx_gb200 (#5602) 
Signed-off-by: bhsueh <11360707+byshiue@users.noreply.github.com>
 2025-07-02 10:07:01 +08:00
Kaiyu Xie
682b164b9b
doc: Fix outdated config in DeepSeek best perf practice doc (#5638) 
Signed-off-by: Kaiyu Xie <26294424+kaiyux@users.noreply.github.com>
 2025-07-01 04:58:50 -04:00
Yi Zhang
aa0b9278d2
test: add more tests for GB200 with 8 GPUs/2 nodes in L0 tests (#5397) 
Signed-off-by: Yi Zhang <187001205+yizhang-nv@users.noreply.github.com>
 2025-07-01 01:06:47 -04:00
Zheng Duan
1824c44004
[nvbug 5300551] test: increase block count in eviction test (#5465) 
Signed-off-by: zhengd-nv <200704041+zhengd-nv@users.noreply.github.com>
 2025-07-01 10:48:25 +08:00
nv-guomingz
9fe1dd6be1
fix:https://nvbugs/5362398 (#5609) 
Signed-off-by: nv-guomingz <137257613+nv-guomingz@users.noreply.github.com>
 2025-06-30 13:29:40 -04:00
Yan Chunwei
d6c81bad97
fix [nvbug5351244]: test_mpi_session submit sync/async (#5608) 
Signed-off-by: Superjomn <328693+Superjomn@users.noreply.github.com>
 2025-07-01 00:48:59 +08:00
Emma Qiao
647e070ed6
[Infra][release/0.21]Update nccl to 2.27.5 (#5539) 
Signed-off-by: qqiao <qqiao@nvidia.com>
 2025-06-29 20:50:15 +08:00
Venky
4fc0666daa
[cherry-pick] [CI] Waive test_fp8_block_scales_4gpus[ep4-mtp_nextn=0-fp8kv=True-attention_dp=True-cuda_graph=True-overlap_scheduler=True-torch_compile=False] (#5553) 
Signed-off-by: Venky <23023424+venkywonka@users.noreply.github.com>
 2025-06-28 01:15:04 +08:00
ixlmar
abb7357f25
[TRTLLM-5989, TRTLLM-5991, TRTLLM-5993] doc: Update container instructions (#5490) 
Signed-off-by: ixlmar <206748156+ixlmar@users.noreply.github.com>
 2025-06-27 07:09:41 -07:00
Yan Chunwei
b78ad754c8
ci: unwaive llmapi launch test (#5281) 
Signed-off-by: Superjomn <328693+Superjomn@users.noreply.github.com>
 2025-06-27 14:10:45 +08:00
Emma Qiao
e2054bb2aa
[Infra][release/0.21] - waive failed tests (#5537) 
Signed-off-by: qqiao <qqiao@nvidia.com>
 2025-06-27 13:58:13 +08:00
ixlmar
312fd47f84
fix: constrain grepping in docker/Makefile (#5493) 
Signed-off-by: ixlmar <206748156+ixlmar@users.noreply.github.com>
 2025-06-26 13:44:40 +02:00
Kaiyu Xie
30a2a8b81c
doc: Fix benchmark cmd in disagg scripts (#5516) 
Signed-off-by: Kaiyu Xie <26294424+kaiyux@users.noreply.github.com>
 2025-06-26 17:23:24 +08:00
ixlmar
a811077f90
fix: fix regression in LOCAL_USER (#5517) 
Signed-off-by: ixlmar <206748156+ixlmar@users.noreply.github.com>
 2025-06-26 11:10:55 +02:00
Anurag Mukkara
c2799d0465
[nvbug/5354825] Fix nougat test image url (#5496) 
Signed-off-by: Anurag Mukkara <134339030+amukkara@users.noreply.github.com>
 2025-06-26 10:10:18 +08:00
Yan Chunwei
87ead4ecbe
[nvbug 5273941] fix: broken cyclic reference detect (#5417) 
Signed-off-by: Superjomn <328693+Superjomn@users.noreply.github.com>
 2025-06-26 07:35:35 +08:00
Martin Marciniszyn Mehringer
fc64f139e4
Fix permission for local user issues in NGC docker container. (#5373) 
Signed-off-by: Martin Marciniszyn Mehringer <11665257+MartinMarciniszyn@users.noreply.github.com>
 2025-06-25 14:10:20 +02:00
Emma Qiao
b6d23d58c4
[Infra] - Waive failed tests on release/0.21 (#5477) 
Signed-off-by: qqiao <qqiao@nvidia.com>
 2025-06-25 19:01:55 +08:00
HuiGao-NV
5cd87bee41
tests: Set kv cache free memory fraction in test case (#5462) 
Signed-off-by: Hui Gao <huig@nvidia.com>
 2025-06-25 16:27:46 +08:00
ruodil
5e50fcc51b
test: set enable_attention_dp=True in default deepseek settings (#5461) 
Signed-off-by: ruodil <200874449+ruodil@users.noreply.github.com>
Co-authored-by: Larry <197874197+LarryXFly@users.noreply.github.com>
 2025-06-25 14:21:14 +08:00
Wanli Jiang
af5839303d
feat: TRTLLM-5941 Upgrade xgrammar to 0.1.18 (#5364) 
Signed-off-by: Wanli Jiang <35160485+Wanli-Jiang@users.noreply.github.com>
 2025-06-25 14:10:50 +08:00
brb-nv
32f50ded17
nvbugs-5331031; nvbugs-5344203 - address intermittent issues with Mistral Small multimodal for BS=8 (#5453) 
Signed-off-by: Balaram Buddharaju <169953907+brb-nv@users.noreply.github.com>
 2025-06-25 11:45:14 +08:00
Ivy Zhang
9e110b2d11
tests: fix typos in qa test (#5421) 
Signed-off-by: Ivy Zhang <25222398+crazydemo@users.noreply.github.com>
 2025-06-25 10:42:34 +08:00
Kaiyu Xie
2b56957fb5
Fix: missing clientId when serialize and deserialize response (cherry-pick #5231) (#5378) 
Signed-off-by: Kaiyu Xie <26294424+kaiyux@users.noreply.github.com>
 2025-06-24 10:00:37 +08:00
Yi Zhang
2d5e202484
fix: Fix skip by mpi size fixture (#5355) 
Signed-off-by: Yi Zhang <187001205+yizhang-nv@users.noreply.github.com>
 2025-06-22 02:51:01 +08:00
Martin Marciniszyn Mehringer
ebc6dbcb0b
doc: cherry pick #5334 (#5368) 
Signed-off-by: Martin Marciniszyn Mehringer <11665257+MartinMarciniszyn@users.noreply.github.com>
 2025-06-19 20:03:59 +08:00
Emma Qiao
8686805a3b
[Infra]cherry pick sanity check yml change for 5080 and 5090 from main (#5363) 
Signed-off-by: qqiao <qqiao@nvidia.com>
 2025-06-19 15:33:57 +08:00
ruodil
e87cf62c12
tests: cherry-pick from main branch, add qwen3 test cases and amend test name in perf test (#5357) 
Signed-off-by: ruodil <200874449+ruodil@users.noreply.github.com>
 2025-06-19 14:34:05 +08:00
Yiqing Yan
decfe2fdb3
chore: bump version to 0.21.0 (#5325) 
Signed-off-by: Yiqing Yan <yiqingy@nvidia.com>
 2025-06-19 12:58:44 +08:00
Yiqing Yan
da576bcafa
Waive L0 test (#5349) 
Signed-off-by: Yiqing Yan <yiqingy@nvidia.com>
 2025-06-19 12:01:11 +08:00
Fanrong Li
6c3210a8be
[test] add nvfp4 DeepSeek-V3-Lite-mtp tests (#5125) 
Signed-off-by: Fanrong Li <23290157+lfr-0531@users.noreply.github.com>
 2025-06-19 09:48:22 +08:00
nv-guomingz
6a388b105a
chore: remove torch_compile prefix for TorchCompileConfig field members (#5261) 
Signed-off-by: nv-guomingz <137257613+nv-guomingz@users.noreply.github.com>
 2025-06-19 09:21:51 +08:00
Zongfei Jing
2b23cd56ce
[feat] Fusion finalize and allreduce for qwenmoe model (#5223) 
Signed-off-by: Zongfei Jing <20381269+zongfeijing@users.noreply.github.com>
Co-authored-by: Kefeng-Duan <176893526+Kefeng-Duan@users.noreply.github.com>
 2025-06-19 08:03:58 +08:00
Robin Kobus
1a7c6e7974
ci: Split long running jobs into multiple jobs (#5268) 
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
Co-authored-by: QI JUN <22017000+QiJune@users.noreply.github.com>
 2025-06-19 06:24:29 +08:00
Yan Chunwei
3946e798db
fix[nvbug5298640]: trtllm-llmapi-launch multiple LLM instances (#4727) 
Signed-off-by: Superjomn <328693+Superjomn@users.noreply.github.com>
 2025-06-19 06:13:53 +08:00