Repository navigation

[None][chore] Mass integration of release/1.0 - 3rd - #7519

Merged

nv-guomingz merged 45 commits into

NVIDIA:mainfrom

dominicshanshan:mi-release-1.0-3

Sep 8, 2025

Merged

[None][chore] Mass integration of release/1.0 - 3rd#7519
nv-guomingz merged 45 commits into
NVIDIA:mainfrom
dominicshanshan:mi-release-1.0-3

Commits on Sep 7, 2025
- [https://nvbugs/5451028][fix] Constrain NemotronSuper test parameters… (#6987)
  
  Naveassaf
  authored and
  dominicshanshan
  committed
  450522e
- [https://nvbugs/5448579][fix] EXAONE-4.0 accuracy test bugfix (#6888)
  
  yechank-nvidia
  authored and
  dominicshanshan
  committed
  ed70d06
- [None][chore] Waive E2E GB200 tests for Gemma3 27B (#6916)
  
  brb-nv
  authored and
  dominicshanshan
  committed
  06c64a7
- [https://nvbugs/5451296][bug] Fix a thread leak in test_llm_args.py (#7017)
  
  Tabrizian
  authored and
  dominicshanshan
  committed
  a60af95
- [None][infra] Waive failed tests for release branch (#7036)
  
  EmmaQiaoCh
  authored and
  dominicshanshan
  committed
  dfbde64
- [None][doc] add status labels to LLM class's api reference (#6899)
  
  Superjomn
  authored and
  dominicshanshan
  committed
  6ca71f5
- [https://nvbugs/5448437][fix] fix some nixl tests (#6940)
  
  bo-nv
  authored and
  dominicshanshan
  committed
  aa6603c
- [https://nvbugs/5427801][fix] Torch compile support for Llama4 and Ea… (#6978)
  
  liji-nv
  authored and
  dominicshanshan
  committed
  b33b27a
- [https://nvbugs/5394392][fix] Enlarge scheduler capacity under disagg bs == 1 (#6975)
  
  yifeizhang-c
  authored and
  dominicshanshan
  committed
  c1e6126
- [TRTLLM-7263][fix] Prevent recreation of cublas handles in lora_grouped_gemm every call (#7053)
  
  amitz-nv
  authored and
  dominicshanshan
  committed
  ede27da
- [None][doc] update v1.0 doc for trtllm-serve (#7056)
  
  hchings
  authored and
  dominicshanshan
  committed
  835192d
- [https://nvbugs/5440241][fix] Fix 70B GSM8K Accuracy drop (#7075)
  
  chenfeiz0326
  authored and
  dominicshanshan
  committed
  e7bc4a6
- [https://nvbugs/5451296][fix] zmq nonblock bug with retry (#7019)
  
  Superjomn
  authored and
  dominicshanshan
  committed
  f8a37bb
- [https://nvbugs/5383702][fix] test_llm_api_pytorch.py::TestLlama3_1_8BInstruct::test_fp8_4gpus (#6889)
  
  Superjomn
  authored and
  dominicshanshan
  committed
  5ff7b61
- [https://nvbugs/5392414] [fix] For release 1.0 cherry pick. Add customized default routing method (#7068)
  
  ChristinaZ
  authored and
  dominicshanshan
  committed
  990a786
- [https://nvbugs/5464088] [fix] dequantize fp8 activation input to lora forward; update perf test config (#7014)
  
  venkywonka
  authored and
  dominicshanshan
  committed
  30f30f7
- [None][infra] Skip failed tests for release branch (#7130)
  
  EmmaQiaoCh
  authored and
  dominicshanshan
  committed
  4b978c8
- [https://nvbugs/5448442][fix] Skip trtllm moe backend for sm120 (#7010)
  
  pamelap-nvidia
  authored and
  dominicshanshan
  committed
  d49e304
- [https://nvbugs/5449032][fix] Add more llm-args to llm_mgmn_trtllm_bench.sh (#7144)
  
  brb-nv
  authored and
  dominicshanshan
  committed
  37823b9
- [https://nvbugs/5410391][bug] Support to share device buffers in attention meta (#6557)
  
  HuiGao-NV
  authored and
  dominicshanshan
  committed
  f4a8c04
- [https://nvbugs/5467062][fix] pass logitsPostProcessorBatched by reference (#7110)
  
  milesial
  authored and
  dominicshanshan
  committed
  c2fecf3
- [https://nvbugs/5450074][fix] Reduce the device memory requirements for testing (#6990)
  
  Shixiaowei02
  authored and
  dominicshanshan
  committed
  c128437
- [https://nvbugs/5433545][fix] TestPhi4MiniInstruct::test_auto_dtype - Use max_seq_len=4096 to fallback to the short RoPE factor (#6895)
  
  moraxu
  authored and
  dominicshanshan
  committed
  5a42ddc
- [https://nvbugs/5448426][fix] Fix illegal memory access in cuda graph (#7127)
  
  peaceh-nv
  authored and
  dominicshanshan
  committed
  b20ea82
- [None][fix] Switch llm api quickstart example location per workflow. (#7182)
  
  nv-guomingz
  authored and
  dominicshanshan
  committed
  633a4d5
- [https://nvbugs/5467232][fix] Fix load_torch_hf_lora to override lora_config.trtllm_modules_to_hf_modules with default only when it has no value (#7168)
  
  Wanli-Jiang
  authored and
  dominicshanshan
  committed
  2a3e17f
- [None][doc] fix tensorrt legacy quickstart page (#7190)
  
  Superjomn
  authored and
  dominicshanshan
  committed
  26dbb32
- [https://nvbugs/5470840][fix] Disaggregated unit test MPI Init handling (#7139)
  
  pcastonguay
  authored and
  dominicshanshan
  committed
  26db89f
- [None][test] add kv cache size in bench metric and fix failed cases (#7211)
  
  ruodil
  authored and
  dominicshanshan
  committed
  3854ef1
- [https://nvbugs/5409416][fix] test_openai_multi_chat_example (#7174)
  
  Linda-Stadter
  authored and
  dominicshanshan
  committed
  ce80090
- [https://nvbugs/5473789][bug] install cuda-toolkit to fix sanity check (#7159)
  
  authored and
  dominicshanshan
  committed
  3af7b1a
Commits on Sep 8, 2025
- [https://nvbugs/5473789][bug] install cuda-toolkit to fix sanity check (#7159)
  
  dominicshanshan
  and
  ZhanruiSunCh
  committed
  7b65fd4
- [None][fix] fix log_once usage (#7210)
  
  yuxianq
  authored and
  dominicshanshan
  committed
  642f622
- [None][infra] Waive failed cases for release/1.0 (#7258)
  
  EmmaQiaoCh
  authored and
  dominicshanshan
  committed
  9165e67
- [https://nvbugs/5451342][fix] Use runtime max_batch_size when cuda_graph_config.max_batch_size is not provided in trtllm-bench (#7031)
  
  jiaganc
  authored and
  dominicshanshan
  committed
  31aaed5
- [None][feat] Skip prefetching consolidated safetensors when appropriate (#7225)
  
  2ez4bz
  authored and
  dominicshanshan
  committed
  1cfb4af
- [https://nvbugs/5430125][ci] Unwaive test case for mistral 3.1 small (#7265)
  
  2ez4bz
  authored and
  dominicshanshan
  committed
  70197dd
- [https://nvbugs/5478151][fix] Add missing spec for Llama-3.3 70B (#7267)
  
  authored and
  dominicshanshan
  committed
  e1d8811
- [https://nvbugs/5451426][fix] Avoid torch compile on full eagle3 worker (#7245)
  
  liji-nv
  authored and
  dominicshanshan
  committed
  413776a
- [https://nvbugs/5463720][fix] tp-split the inferred `mlp_hidden_size` for nemotron-nas (#7231)
  
  venkywonka
  authored and
  dominicshanshan
  committed
  7ff6f44
- [https://nvbugs/5480550][fix] Increase timeout for Gemma3 27B test (#7271)
  
  brb-nv
  authored and
  dominicshanshan
  committed
  cc71861
- [https://nvbugs/5434320][bug] Fix disagg pp bug (#7099)
  
  Tabrizian
  authored and
  dominicshanshan
  committed
  8797444
- [https://nvbugs/5480415][fix] Fix phi4mm multi-gpu test (#7275)
  
  authored and
  dominicshanshan
  committed
  d279d29
- [TRTLLM-7346][fix] Improve performance of PyTorchModelEngine._get_lora_params_from_requests (#7203)
  
  amitz-nv
  authored and
  dominicshanshan
  committed
  2083332
- [https://nvbugs/5461712] [fix] Disable deep_gemm for Qwen3 due to accuracy issues (#7170)
  
  DomBrown
  authored and
  dominicshanshan
  committed
  74fc47c

Commits

Commits on Sep 7, 2025

Commits on Sep 8, 2025