Commits · f0ca0671c70fae6d1562127e3330eeaedf4abb3f · OpenDAS / vllm_cscc · GitLab

10 Feb, 2026 40 commits
- [Feature] Warn about unrecognized environment variables (#33581) · f0ca0671
  Gregory Shtrasberg authored Feb 10, 2026
```
Signed-off-by: Gregory Shtrasberg <Gregory.Shtrasberg@amd.com>
```
  f0ca0671
- [SM100] Resubmit FMHA FP8 prefill for MLA (#31195) · 578977bb
  Pavani Majety authored Feb 10, 2026
```
Signed-off-by: Pavani Majety <pmajety@nvidia.com>
```
  578977bb
- [Bugfix] Fix mamba cache dtype for Qwen3.5 (#34200) · 9615575a
  Roger Wang authored Feb 10, 2026
```
Signed-off-by: Roger Wang <hey@rogerw.io>
```
  9615575a
- [Benchmarks] Fix attention benchmark smoke test (#34269) · 4293c00b
  Matthew Bonanni authored Feb 10, 2026
```
Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>
```
  4293c00b
- [Bugfix] Fix weights offloading for sleep mode (#32947) · 506ad7d7
  J Seppänen authored Feb 10, 2026
```
Signed-off-by: Jarno Seppänen <jseppanen@nvidia.com>
Co-authored-by: Tyler Michael Smith <tyler@neuralmagic.com>
```
  506ad7d7
- Convert online APIs to use Renderer (#34084) · fdd6f2ad
  Reagan Lee authored Feb 10, 2026
```
Signed-off-by: Reagan Lee <“reaganjlee@gmail.com”>
Co-authored-by: Reagan Lee <“reaganjlee@gmail.com”>
```
  fdd6f2ad
- [Misc] Introduce ec_both role EC (encoder cache) connector (#34182) · 33bcd3dc
  Qi Wang authored Feb 10, 2026
```
Signed-off-by: Qi Wang <qiwa@nvidia.com>
```
  33bcd3dc
- [UX nit] Fix non-default api_server_count message (#34152) · 1f5febb4
  Michael Goin authored Feb 10, 2026
```
Signed-off-by: mgoin <mgoin64@gmail.com>
```
  1f5febb4
- Minor cleanup for Voxtral (#34247) · ae871ca9
  Andy Lo authored Feb 10, 2026
```
Signed-off-by: Andy Lo <andy@mistral.ai>
```
  ae871ca9
- [Model Runner V2] Use pinned memory for write_contents (#34222) · a2443de5
  Woosuk Kwon authored Feb 10, 2026
```
Signed-off-by: Woosuk Kwon <woosuk@inferact.ai>
```
  a2443de5
- [Docs] Speed up build environment set-up (#34240) · f84a2a8f
  Harry Mellor authored Feb 10, 2026
```
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
```
  f84a2a8f
- [BUGFIX] Fix accuracy bugs in Qwen3-Next MTP (#34077) · 000214c4
  Vadim Gimpelson authored Feb 10, 2026
```
Signed-off-by: Vadim Gimpelson <vadim.gimpelson@gmail.com>
```
  000214c4
- [Core][BugFix] Fix PP KV cache sharding memory validation (#33698) · c5a66d16
  junuxyz authored Feb 11, 2026
```
Signed-off-by: junuxyz <216036880+junuxyz@users.noreply.github.com>
```
  c5a66d16
- [Perf][Kernel] Add faster topKperRow decode kernel for DeepSeek-V3.2 sparse attention (#33680) · afdce12c
  Roberto L. Castro authored Feb 10, 2026
```
Signed-off-by: LopezCastroRoberto <rocastro@redhat.com>
Signed-off-by: Roberto L. Castro <38211239+LopezCastroRoberto@users.noreply.github.com>
Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
```
  afdce12c
- [compile] Enable AOT compile with 2.10 in trunk. (#34155) · 82e11973
  Zhengxu Chen authored Feb 10, 2026
```
Signed-off-by: Zhengxu Chen <zhxchen17@meta.com>
```
  82e11973
- [ROCm][Quantization] GPT_OSS in amd-quark format model loading and emulations (#29008) · b129136c
  xuebwang-amd authored Feb 10, 2026
```
Signed-off-by: xuebwang-amd <xuebwang@amd.com>
Signed-off-by: Robert Shaw <robshaw@redhat.com>
Co-authored-by: Robert Shaw <robshaw@redhat.com>
Co-authored-by: Robert Shaw <114415538+robertgshaw2-redhat@users.noreply.github.com>
```
  b129136c
- Support benchmarking of Geospatial models (#33922) · 599e4335
  mgazz authored Feb 10, 2026
```
Signed-off-by: Michele Gazzetti <michele.gazzetti1@ibm.com>
```
  599e4335
- add --insecure arg to the vllm bench to skip TLS (#34026) · a1946570
  Fan Yang authored Feb 10, 2026
```
Signed-off-by: Fan Yang <yan9fan@meta.com>
Co-authored-by: Fan Yang <yan9fan@meta.com>
```
  a1946570
- Bump `mamba-ssm` version in CI for Transformers v5 compatibility (#34233) · d0bc5205
  Harry Mellor authored Feb 10, 2026
```
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
```
  d0bc5205
- [V1][BugFix] Fix EAGLE3 encoder cache miss with disable_chunked_mm_input (#34220) · 748625cd
  Krish Gupta authored Feb 10, 2026
```
Signed-off-by: KrxGu <krishom70@gmail.com>
```
  748625cd
- Stop testing for slow tokenizers as they will not exist soon (#34235) · 61413973
  Harry Mellor authored Feb 10, 2026
```
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
```
  61413973
- [Misc] allow specify is_mm_prefix_lm in hf_config (#34215) · 94de8715
  Phúc H. Lê Khắc authored Feb 10, 2026
  
  94de8715
- Add flagos in MiniCPM-o (#34126) · e042d7e6
  tc-mb authored Feb 10, 2026
```
Signed-off-by: tc-mb <caitianchi@modelbest.cn>
Signed-off-by: Vincent-Xiao <vincent.xiao.me@gmail.com>
Co-authored-by: Vincent-Xiao <vincent.xiao.me@gmail.com>
```
  e042d7e6
- [Bugfix] Fix FI kernel`chunk_gated_delta_rule` output shape for Qwen3.5 (#34219) · ae4e2806
  Roger Wang authored Feb 10, 2026
```
Signed-off-by: Roger Wang <hey@rogerw.io>
```
  ae4e2806
- [Docs] Fix format error in KV load failure recovery doc (#34137) · cbea11c9
  zzaebok authored Feb 10, 2026
```
Signed-off-by: Jaebok Lee <jaebok9541@naver.com>
```
  cbea11c9
- [Bugfix] Fix `--trust-remote-code` conflict (#34218) · 2c32558a
  Cyrus Leung authored Feb 10, 2026
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  2c32558a
- [Bugfix] Fix memory inconsistency in cross-process shared memory (#32022) · 5f970120
  Zetong Li authored Feb 10, 2026
```
Signed-off-by: Zetong Li <slippersss@126.com>
```
  5f970120
- Revert #34208 (#34216) · 998e2d91
  Cyrus Leung authored Feb 10, 2026
  
  998e2d91
- [Perf] Optimize detokenizer python logic (#32975) · e1060a71
  Wentao Ye authored Feb 10, 2026
```
Signed-off-by: yewentao256 <zhyanwentao@126.com>
Signed-off-by: Wentao Ye <44945378+yewentao256@users.noreply.github.com>
Co-authored-by: Nick Hill <nhill@redhat.com>
```
  e1060a71
- [BugFix] Avoid prefix cache hit in the same schedule step for mamba layers (#29387) · 97fa8f65
  Chen Zhang authored Feb 09, 2026
```
Signed-off-by: Chen Zhang <zhangch99@outlook.com>
```
  97fa8f65
- [Frontend][CI] Consolidate instrumentator entrypoints (#34123) · dab1de9f
  wang.yuqi authored Feb 10, 2026
```
Signed-off-by: wang.yuqi <yuqi.wang@daocloud.io>
```
  dab1de9f
- [Bugfix] Sort hf_weights_files in fastsafetensors_weights_iterator to match #33491 (#34190) · 8d48d0a9
  Balaxxe authored Feb 10, 2026
```
Signed-off-by: Balaxxe <136368465+jaim12005@users.noreply.github.com>
```
  8d48d0a9
- [responsesAPI] fix simpleContext streaming output_messages (#34188) · 9608844f
  Andrew Xia authored Feb 09, 2026
```
Signed-off-by: Andrew Xia <axia@meta.com>
Signed-off-by: Andrew Xia <axia@fb.com>
Co-authored-by: Andrew Xia <axia@fb.com>
```
  9608844f
- [Bugfix] Add `--trust-remote-code` to dataset bench args (#34208) · f69b903b
  Cyrus Leung authored Feb 10, 2026
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  f69b903b
- [Bugfix] Fix DP Attention Padding in Dummy Run (#34187) · 81e217fe
  Lucas Wilkinson authored Feb 09, 2026
```
Signed-off-by: Benjamin Chislett <bchislett@nvidia.com>
Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
Co-authored-by: Benjamin Chislett <bchislett@nvidia.com>
```
  81e217fe
- [CI/Build] Relax `test_mcp_tool_call` (#34204) · ab97bcf6
  Cyrus Leung authored Feb 10, 2026
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  ab97bcf6
- [Doc] Update usage of `--limit-mm-per-prompt` (#34148) · 25e48a3a
  Cyrus Leung authored Feb 10, 2026
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  25e48a3a
- [Bugfix][Core] Fix CPU memory leak from Request reference cycle in prefix caching (#34183) · 8a5e0e2b
  Roger Wang authored Feb 09, 2026
```
Signed-off-by: Roger Wang <hey@rogerw.io>
```
  8a5e0e2b
- [ROCm][Bugfix] Resolve Dynamo tracing crash from amdsmi calls in on_gfx* arch detection (#34108) · 4cde2e01
  Andreas Karatzas authored Feb 09, 2026
```
Signed-off-by: Andreas Karatzas <akaratza@amd.com>
```
  4cde2e01
- [Bugfix] Adopt `ChunkGatedDeltaRule` for Qwen3.5 (#34198) · 047a457f
  Roger Wang authored Feb 09, 2026
```
Signed-off-by: Roger Wang <hey@rogerw.io>
```
  047a457f