Commits · 19cc9468fd0fa1701e7cb74b5928b329a1d16cf1 · OpenDAS / vllm_cscc

22 Dec, 2025 1 commit
- [Feature]: Support NVIDIA ModelOpt HF FP8 variants FP8_PER_CHANNEL_PER_TOKEN... · 19cc9468
  CedricHuang authored Dec 22, 2025
```
[Feature]: Support NVIDIA ModelOpt HF FP8 variants FP8_PER_CHANNEL_PER_TOKEN and FP8_PB_WO  in vLLM (#30957)
```
  19cc9468
21 Dec, 2025 2 commits

[MoE Refactor][4/N] Marlin Fp8 Mk (#31036) · b471092d
Robert Shaw authored Dec 21, 2025

b471092d

[BugFix]fix gpt-oss v1/completions response bug (#30608) · 3e92b2b7

汪志鹏 authored Dec 21, 2025


Signed-off-by: princepride <wangzhipeng628@gmail.com>
Signed-off-by: 汪志鹏 <wangzhipeng628@gmail.com>
Co-authored-by: Chauncey <chaunceyjiang@gmail.com>
Co-authored-by: Cyrus Leung <tlleungac@connect.ust.hk>
Co-authored-by: bbrowning <bbrownin@redhat.com>

3e92b2b7

20 Dec, 2025 7 commits

[Quantization] add marlin w4a8/w8a8 check (#31061) · 7c73ceb5
Jinzhen Lin authored Dec 21, 2025
```
Signed-off-by: Jinzhen Lin <jinzhen.ljz@antgroup.com>
```
7c73ceb5

[Quantization] support logical_widths for fp8 marlin (#30962) · ee52d990

Jinzhen Lin authored Dec 21, 2025

Signed-off-by: Jinzhen Lin <jinzhen.ljz@antgroup.com>
Signed-off-by: Jinzhen Lin <linjinzhen@hotmail.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>

ee52d990

[MoE Refactor][5/N] Isolate zero expert to LongCatFlash (#28891) · 54c89243

baonudesifeizhai authored Dec 20, 2025

Signed-off-by: baonudesifeizhai <85092850+baonudesifeizhai@users.noreply.github.com>
Signed-off-by: Dongjie Zou <85092850+baonudesifeizhai@users.noreply.github.com>
Signed-off-by: baonudesifeizhai <baonudesifeizhai@gmail.com>
Signed-off-by: Robert Shaw <robertgshaw2@gmail.com>
Co-authored-by: Robert Shaw <robshaw@redhat.com>
Co-authored-by: Robert Shaw <robertgshaw2@gmail.com>

54c89243

[XPU] enable fp8 online streaming quantization (#30944) · 560ae963
Yan Ma authored Dec 20, 2025
```
Signed-off-by: Yan Ma <yan.ma@intel.com>
```
560ae963
[Bugfix] Read truncate_prompt_tokens from pooling_params in AsyncLLM.encode() (#31013) · 1501a407
Jeffrey Wang authored Dec 20, 2025
```
Signed-off-by: Jeffrey Wang <jeffreywang@anyscale.com>
```
1501a407

[Bugfix] fix the alias bug of AttentionBackendEnum when register CUSTOM... · d52c5096

zejunchen-zejun authored Dec 20, 2025


[Bugfix] fix the alias bug of AttentionBackendEnum when register CUSTOM attention backend to vllm (#30869)
Signed-off-by: zejunchen-zejun <zejun.chen@amd.com>

d52c5096

GLM-4.7 Tool Parser and Doc Update (#30876) · 8a7a4143
Yuxuan Zhang authored Dec 20, 2025
```
Signed-off-by: zRzRzRzRzRzRzR <2448370773@qq.com>
```
8a7a4143

19 Dec, 2025 20 commits
- [MoE Refactor][2/N] Use Modular Kernels for Fp8 (#30825) · 95befecc
  Robert Shaw authored Dec 19, 2025
```
Signed-off-by: Robert Shaw <robshaw@redhat.com>
Co-authored-by: Robert Shaw <robshaw@redhat.com>
```
  95befecc
- [Bug] Fix `error 'Dynamo failed to run FX node with fake tensors` for Deepseek V3.2 (#31046) · 4cf94298
  Wentao Ye authored Dec 19, 2025
```
Signed-off-by: yewentao256 <zhyanwentao@126.com>
```
  4cf94298
- [MoE Refactor][3/N] Deprecate cutlass block quant fp8 (b200) (#30990) · 83a317f6
  Robert Shaw authored Dec 19, 2025
```
Signed-off-by: Robert Shaw <robshaw@redhat.com>
Co-authored-by: Robert Shaw <robshaw@redhat.com>
```
  83a317f6
- [BugFix] Fix TypeError: unhashable type: 'dict' when serving deepseek32 (#30924) · 5f6477d1
  Lucas Wilkinson authored Dec 19, 2025
```
Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
```
  5f6477d1
- [Refactor] Refactor for `DeepGemmQuantScaleFMT` using cache (#30898) · 3bd8335b
  Wentao Ye authored Dec 19, 2025
```
Signed-off-by: yewentao256 <zhyanwentao@126.com>
```
  3bd8335b
- Make engine core client handshake timeout configurable (#27444) · 1ab52135
  Seiji Eicher authored Dec 19, 2025
```
Signed-off-by: Seiji Eicher <seiji@anyscale.com>
```
  1ab52135
- [Model] Add MiMo-V2-Flash support (#30836) · 969bbc7c
  Zhonghua Deng authored Dec 20, 2025
```
Signed-off-by: Abatom <abzhonghua@gmail.com>
Signed-off-by: Jumiar <liuanqim10@126.com>
Signed-off-by: Zyann7 <zyann7@outlook.com>
Co-authored-by: Jumiar <liuanqim10@126.com>
Co-authored-by: Zyann7 <zyann7@outlook.com>
Co-authored-by: Jee Jee Li <pandaleefree@gmail.com>
```
  969bbc7c
- [Quantization] fix marlin w8a8 check (#30961) · 5fbfa8d9
  Jinzhen Lin authored Dec 19, 2025
```
Signed-off-by: Jinzhen Lin <jinzhen.ljz@antgroup.com>
```
  5fbfa8d9
- [CustomOp][Refactor] Extract common methods for ApplyRotaryEmb CustomOp (#31021) · 23a1946e
  Shanshan Shen authored Dec 19, 2025
```
Signed-off-by: shen-shanshan <467638484@qq.com>
```
  23a1946e
- [Bugfix] [Kernel] Triton attention kernels: mask out V blocks that fall... · b5545d9d
  Thomas Parnell authored Dec 19, 2025
```
[Bugfix] [Kernel] Triton attention kernels: mask out V blocks that fall outside sliding window (#30887)
Signed-off-by: Thomas Parnell <tpa@zurich.ibm.com>
```
  b5545d9d
- [Frontend][Bug] allow tool calls in analysis channel (#28139) · 45594967
  Marko Rosenmueller authored Dec 19, 2025
```
Signed-off-by: Marko Rosenmueller <5467316+dr75@users.noreply.github.com>
Co-authored-by: Chauncey <chaunceyjiang@gmail.com>
```
  45594967
- [Bugfix] Add validation for tool requests when tool_parser is unavailable (#30613) · 086b9633
  lif authored Dec 19, 2025
```
Signed-off-by: majiayu000 <1835304752@qq.com>
Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>
```
  086b9633
- [Quantization] enable compressed-tensors marlin support for turing (2) (#31008) · 9187de9f
  Jinzhen Lin authored Dec 19, 2025
```
Signed-off-by: Jinzhen Lin <jinzhen.ljz@antgroup.com>
```
  9187de9f
- [Bugfix] Fix incorrect tiles creation for mm prefix triton attention (#30974) · ac1c9342
  Isotr0py authored Dec 19, 2025
```
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
```
  ac1c9342
- Add hidden dimension validation for multimodal embedding inputs (#30968) · 4924ac58
  Wenqi Glantz authored Dec 19, 2025
```
Signed-off-by: Wenqi Glantz <wglantz@nvidia.com>
```
  4924ac58
- [Quantization] enable compressed-tensors marlin support for turing (#31000) · de08b8f6
  Jinzhen Lin authored Dec 19, 2025
```
Signed-off-by: Jinzhen Lin <jinzhen.ljz@antgroup.com>
```
  de08b8f6
- [BugFix] Fix logprobs with spec decode and modified logits (#30846) · 2ac85a45
  Nick Hill authored Dec 18, 2025
```
Signed-off-by: Nick Hill <nhill@redhat.com>
```
  2ac85a45
- [ROCm][CI][Bugfix] Multi-Modal Model Support Fixes and Attention Backend Improvements (#30270) · 7b43db21
  Andreas Karatzas authored Dec 18, 2025
```
Signed-off-by: Andreas Karatzas <akaratza@amd.com>
```
  7b43db21
- [Bugfix] Fix tool_choice="none" being ignored by GPT-OSS/harmony models (#30867) · 6a09612b
  PlatinumGod authored Dec 19, 2025
```
Signed-off-by: yujiepu <pyjapple@gmail.com>
Signed-off-by: PlatinumGod <pyjapple@gmail.com>
Co-authored-by: Chauncey <chaunceyjiang@gmail.com>
```
  6a09612b
- [BugFix] Handle errors when preprocessing added requests (#30895) · 45c0526a
  Nick Hill authored Dec 18, 2025
```
Signed-off-by: Nick Hill <nhill@redhat.com>
```
  45c0526a
18 Dec, 2025 10 commits
- [Cleanup] Refactor FlashInferMetadataBuilder (#29128) · d6b3d39b
  Benjamin Chislett authored Dec 18, 2025
```
Signed-off-by: Benjamin Chislett <bchislett@nvidia.com>
Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
Co-authored-by: Lucas Wilkinson <lwilkins@redhat.com>
```
  d6b3d39b
- [NIXL][BUG FIX] Fix both failing issue and accuracy issue with nixl + host_buffer on CUDA (#30419) · 6ca74bc1
  Chendi.Xue authored Dec 18, 2025
```
Signed-off-by: Chendi Xue <chendi.xue@intel.com>
Signed-off-by: Chendi.Xue <chendi.xue@intel.com>
Co-authored-by: Nicolò Lucchesi <nlucches@redhat.com>
```
  6ca74bc1
- Check for truthy `rope_parameters` not the existence of it (#30983) · 19c58339
  Harry Mellor authored Dec 18, 2025
```
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
```
  19c58339
- [BugFix] Fix spec decode + structured outputs + preemption edge case (#30916) · b0b77c46
  Nick Hill authored Dec 18, 2025
```
Signed-off-by: Nick Hill <nhill@redhat.com>
```
  b0b77c46
- Strengthen input validation and tests for 'parse_raw_prompts’. (#30652) · 634a14bd
  Kayvan Mivehnejad authored Dec 18, 2025
```
Signed-off-by: Kayvan Mivehnejad <K.Mivehnejad@gmail.com>
```
  634a14bd
- [BugFix] Spec decode with VLLM_ENABLE_V1_MULTIPROCESSING=0 (#30319) · 24b65eff
  Chen Zhang authored Dec 18, 2025
```
Signed-off-by: Chen Zhang <zhangch99@outlook.com>
```
  24b65eff
- Remove all2all backend envvar (#30363) · 41b6f920
  Elizabeth Thomas authored Dec 18, 2025
```
Signed-off-by: Elizabeth Thomas <email2eliza@gmail.com>
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
Co-authored-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
```
  41b6f920
- [Bug] Fix compressed tensor not using deepgemm (#30820) · 97000a2b
  Wentao Ye authored Dec 18, 2025
```
Signed-off-by: yewentao256 <zhyanwentao@126.com>
```
  97000a2b
- [Bugfix] Remove `tile_size=64` for mm_prefix triton attention (#30973) · d2dc5dfc
  Isotr0py authored Dec 19, 2025
```
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
```
  d2dc5dfc
- tuned fused configs for B300 (#30629) · b8c477c1
  navmarri14 authored Dec 18, 2025
  
  b8c477c1