Commits · f120bd42d3daf733425d7feaaeffc2a23ba71c17 · OpenDAS / vllm_cscc

12 Feb, 2026 15 commits
- [Kernel] Support Flashinfer trtllm fused MoE non gated FP8 & NVFP4 (#33506) · f120bd42
  amitz-nv authored Feb 12, 2026
```
Signed-off-by: amitz-nv <203509407+amitz-nv@users.noreply.github.com>
```
  f120bd42
- [Bugfix] Enforce DeepGEMM when using sparse_attn_indexer on CUDA (#34374) · 6d4e27ce
  Michael Goin authored Feb 12, 2026
```
Signed-off-by: mgoin <mgoin64@gmail.com>
```
  6d4e27ce
- [Voxtral Realtime] Refactor & Improve buffering logic (#34428) · 6c0baee6
  Patrick von Platen authored Feb 12, 2026
```
Signed-off-by: Patrick von Platen <patrick.v.platen@gmail.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
```
  6c0baee6
- [Voxstral Realtime] Enable tests (#33803) · 1100a976
  Patrick von Platen authored Feb 12, 2026
```
Signed-off-by: Patrick von Platen <patrick.v.platen@gmail.com>
```
  1100a976
- [ROCm][quantization] improve OCP weight quant parser robust (#34431) · 766e1678
  xuebwang-amd authored Feb 13, 2026
```
Signed-off-by: xuebwang-amd <xuebwang@amd.com>
Co-authored-by: TJian <tunjian.tan@embeddedllm.com>
```
  766e1678
- [Bugfix] Remove broken raw url GGUF model loading support (#34433) · becbe248
  Isotr0py authored Feb 13, 2026
```
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
```
  becbe248
- Fix MoE for the Transformers modelling backend (#34436) · 679ca5d8
  Harry Mellor authored Feb 12, 2026
```
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
```
  679ca5d8
- [Attention] Add FlashInfer Sparse MLA backend (#33451) · f2c47886
  Matthew Bonanni authored Feb 12, 2026
```
Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>
Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
Co-authored-by: Lucas Wilkinson <lwilkins@redhat.com>
Co-authored-by: Lucas Wilkinson <LucasWilkinson@users.noreply.github.com>
```
  f2c47886
- Add config file for fused MoE for Nemotron (TP4, B200) (#34411) · dea63512
  danisereb authored Feb 12, 2026
```
Signed-off-by: Daniel Serebrenik <daserebrenik@nvidia.com>
```
  dea63512
- [ROCm] Enable MXFP4 MoE weight pre-shuffling on gfx950 and update aiter (#34192) · 8a798be9
  Douglas Lehr authored Feb 12, 2026
```
Signed-off-by: Doug Lehr <douglehr@amd.com>
Co-authored-by: Doug Lehr <douglehr@amd.com>
Co-authored-by: Gregory Shtrasberg <156009573+gshtras@users.noreply.github.com>
Co-authored-by: tjtanaavllm <tunjian.tan@amd.com>
```
  8a798be9
- [bugfix] refactor FunASR's _get_data_parser (#34397) · 386bfe5d
  AllenDou authored Feb 12, 2026
```
Signed-off-by: zixiao <shunli.dsl@alibaba-inc.com>
Co-authored-by: zixiao <shunli.dsl@alibaba-inc.com>
```
  386bfe5d
- [Bugfix] Fix Sparse24 Compressed Tensors models (#33446) · e9cd6911
  Kyle Sayers authored Feb 12, 2026
```
Signed-off-by: Kyle Sayers <kylesayrs@gmail.com>
Co-authored-by: Michael Goin <mgoin64@gmail.com>
```
  e9cd6911
- Fix DeepSeek-OCR tensor validation for all size variants (#34085) · 80f2ba6e
  Yichuan Wang authored Feb 11, 2026
```
Co-authored-by: Cursor <cursoragent@cursor.com>
```
  80f2ba6e
- [BugFix] Fix DP chunking (#34379) · 136b0bfa
  Lucas Wilkinson authored Feb 11, 2026
```
Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
Signed-off-by: Bill Nell <bnell@redhat.com>
Co-authored-by: Bill Nell <bnell@redhat.com>
```
  136b0bfa
- [Refactor] Replace `activation: str` with `MoEActivation` enum (#33843) · ff1f83b0
  Michael Goin authored Feb 11, 2026
```
Signed-off-by: mgoin <mgoin64@gmail.com>
Signed-off-by: Michael Goin <mgoin64@gmail.com>
```
  ff1f83b0
11 Feb, 2026 19 commits
- [Bugfix] Fix some issues with MoERunner PR #32344 (#34371) · 31d992d2
  bnellnm authored Feb 11, 2026
```
Signed-off-by: Bill Nell <bnell@redhat.com>
```
  31d992d2
- [Bugfix] Fix more multimodal tests for transformers V5 (#34334) · 527ca321
  Raushan Turganbay authored Feb 11, 2026
```
Signed-off-by: raushan <raushan@huggingface.co>
```
  527ca321
- [GPT-OSS] Remove unnecessary contiguous (#34337) · 83e26c83
  elvischenv authored Feb 12, 2026
```
Signed-off-by: elvischenv <219235043+elvischenv@users.noreply.github.com>
```
  83e26c83
- [Bugfix] Enable attn quantization of Llama-4 by correctly permuting scales for... · 11c7ace3
  Eldar Kurtić authored Feb 11, 2026
```
[Bugfix] Enable attn quantization of Llama-4 by correctly permuting scales for rope (int8, fp8) (#34243)
Signed-off-by: Your Name <you@example.com>
Co-authored-by: Your Name <you@example.com>
```
  11c7ace3
- [Bugfix] fix default is_neox_style is True for deepseek (#34353) · be7f3d5d
  Xinyu Dong authored Feb 12, 2026
```
Signed-off-by: dongxinyu03 <dongxinyu03@baidu.com>
```
  be7f3d5d
- [Multimodal] Expose `mm_processor_kwargs` for `DummyInputsBuilder` (#34330) · 0ab06100
  Isotr0py authored Feb 12, 2026
```
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
```
  0ab06100
- [NVIDIA][test] Tests for flashinfer TRTLLM BF16 MoE (#33715) · 275e0d2a
  Linda authored Feb 11, 2026
```
Signed-off-by: Linda-Stadter <57756729+Linda-Stadter@users.noreply.github.com>
Co-authored-by: Pavani Majety <pmajety@nvidia.com>
```
  275e0d2a
- Make JAIS compatible with Transformers v5 (#34264) · 0f5e55e7
  Harry Mellor authored Feb 11, 2026
```
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
```
  0f5e55e7
- Make Qwen3VL compatible with Transformers v5 (#34262) · 1e9204bf
  Harry Mellor authored Feb 11, 2026
```
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
Signed-off-by: Roger Wang <hey@rogerw.io>
Co-authored-by: Roger Wang <hey@rogerw.io>
```
  1e9204bf
- [Bugfix][CPU] Fix llama4 inference on CPU (#34321) · 05339a7b
  Li, Jiang authored Feb 11, 2026
```
Signed-off-by: jiang1.li <jiang1.li@intel.com>
```
  05339a7b
- [Docs] Reduce time spent generating API docs (#34255) · 40b8f553
  Harry Mellor authored Feb 11, 2026
```
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
```
  40b8f553
- [XPU][9/N] clean up existing ipex code/doc (#34111) · cb9574eb
  Kunshang Ji authored Feb 11, 2026
```
Signed-off-by: Kunshang Ji <kunshang.ji@intel.com>
```
  cb9574eb
- [model] support FunASR model (#33247) · 21dfb842
  AllenDou authored Feb 11, 2026
```
Signed-off-by: zixiao <shunli.dsl@alibaba-inc.com>
Co-authored-by: zixiao <shunli.dsl@alibaba-inc.com>
```
  21dfb842
- [Bugfix] Fix weight naming in Qwen3.5 (#34313) · 0b20469c
  Roger Wang authored Feb 10, 2026
```
Signed-off-by: Roger Wang <hey@rogerw.io>
```
  0b20469c
- [Bugfix] Fix fused MoE IMA (sans chunking) by using int64 for strides (#34279) · d7982daf
  Tyler Michael Smith authored Feb 11, 2026
```
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
```
  d7982daf
- [ModelBash][DSR1 NVFp4] Removed Bf16 Bias Cast (#34298) · 9b17c574
  Robert Shaw authored Feb 11, 2026
```
Signed-off-by: Robert Shaw <robshaw@redhat.com>
Co-authored-by: Robert Shaw <robshaw@redhat.com>
```
  9b17c574
- [XPU][7/N] enable xpu fp8 moe (#34202) · b482f71e
  zofia authored Feb 11, 2026
```
Signed-off-by: Zhu, Zufang <zufang.zhu@intel.com>
```
  b482f71e
- [WideEP] Fix nvfp4 DeepEP High Throughput All2All backend (#33738) · 066c6da6
  Tyler Michael Smith authored Feb 10, 2026
```
Signed-off-by: Tyler Michael Smith <tlrmchlsmth@gmail.com>
Co-authored-by: Robert Shaw <114415538+robertgshaw2-redhat@users.noreply.github.com>
```
  066c6da6
- [MoE Refactor] Introduce MoERunner abstraction and move execution logic from... · d1481ba7
  bnellnm authored Feb 10, 2026
```
[MoE Refactor] Introduce MoERunner abstraction and move execution logic from FusedMoE to DefaultMoERunner (#32344)
Signed-off-by: Bill Nell <bnell@redhat.com>
```
  d1481ba7
10 Feb, 2026 6 commits

[Misc][Spec Decode] support different load config for draft model (#34022) · 6f2f59f2

Zhengkai Zhang authored Feb 10, 2026


Signed-off-by: zzhengkai <zzhengkai@devgpu049.ldc1.facebook.com>
Co-authored-by: zzhengkai <zzhengkai@devgpu049.ldc1.facebook.com>

6f2f59f2

[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
Minor cleanup for Voxtral (#34247) · ae871ca9
Andy Lo authored Feb 10, 2026
```
Signed-off-by: Andy Lo <andy@mistral.ai>
```
ae871ca9

[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

[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