Commits · d468322dc1c89377f65488973d2f1927e9bbe5cb · OpenDAS / vllm_cscc

10 Apr, 2026 8 commits

[Kernel][Hardware][AMD] Add TritonW4A16LinearKernel for ROCm (#37352) · d468322d

jatseng-ai authored Apr 10, 2026


Signed-off-by: jatseng-ai <jatseng@amd.com>
Signed-off-by: jatseng-ai <janet.tseng@amd.com>
Signed-off-by: Matthias Gehre <matthias.gehre@amd.com>
Co-authored-by: Claude <noreply@anthropic.com>
Co-authored-by: Matthias Gehre <matthias.gehre@amd.com>

d468322d

[model] support FireRedLID (#39290) · 967146e7

PatchyTIS authored Apr 10, 2026


Signed-off-by: PatchouliTaisa <patchychen@tencent.com>
Co-authored-by: PatchouliTaisa <patchychen@tencent.com>

967146e7

[ZenCPU] Make PT Backport Patch Accessible to vLLM (#38205) · 8e8a3bec

Lalithnarayan C authored Apr 10, 2026

Signed-off-by: Lalithnarayan C <Lalithnarayan.C@amd.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
Co-authored-by: Luka Govedič <ProExpertProg@users.noreply.github.com>

8e8a3bec

[compile] Allow strings in custom ops without regressing compilation times (#38123) · f44afef6
Richard Zou authored Apr 10, 2026
```
Signed-off-by: Richard Zou <zou3519@gmail.com>
```
f44afef6
[GGUF] Support non-standard quant types with prefix (e.g. UD-IQ1_S) (#39471) · 447ce222
Injae Ryou authored Apr 10, 2026
```
Signed-off-by: Injae Ryou <injaeryou@gmail.com>
Co-authored-by: Isotr0py <mozf@mail2.sysu.edu.cn>
```
447ce222
Add platform manual_seed_all API (#38468) · ec68d53b
Yan Ma authored Apr 10, 2026
```
Signed-off-by: Yan Ma <yan.ma@intel.com>
```
ec68d53b

Add EXAONE-4.5 (#39388) · e7a1387e

Kyungmin Lee authored Apr 10, 2026


Signed-off-by: lkm2835 <lkm2835@gmail.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>

e7a1387e

[Feature] Add auto-detection for reasoning_config when only reasoning_parser is set (#38214) · ecbfbb8d
Chauncey authored Apr 10, 2026
```
Signed-off-by: chaunceyjiang <chaunceyjiang@gmail.com>
```
ecbfbb8d

09 Apr, 2026 13 commits

[ASR] Fix spacing bw chunks in multi chunk audio transcription (#39116) · f7cad674
Ekagra Ranjan authored Apr 09, 2026
```
Signed-off-by: Ekagra Ranjan <3116519+ekagra-ranjan@users.noreply.github.com>
```
f7cad674
[Quantization] Support Quark W8A8 INT8 MoE inference (#36320) · 827268e9
PikaPikachu authored Apr 10, 2026
```
Signed-off-by: kangletian <Letian.Kang@amd.com>
```
827268e9

[Performance Improvement] Update `batched_count_greater_than` to handle batch... · a8c6ee9b

Lucas Kabela authored Apr 09, 2026


[Performance Improvement] Update `batched_count_greater_than` to handle batch size 1 without recompile (#38933)
Signed-off-by: Lucas Kabela <lucaskabela@meta.com>
Co-authored-by: Luka Govedič <ProExpertProg@users.noreply.github.com>

a8c6ee9b

[CI/Build] Fix memory cleanup in MM test (#39411) · 3b1d9c31
Cyrus Leung authored Apr 09, 2026
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
3b1d9c31

[Tests] Add Qwen3-VL multimodal memory leak check (#39268) · 91eea723

lalit10 authored Apr 09, 2026


Signed-off-by: Lalit Laxminarayan Bangad <lalitbangad@gmail.com>
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
Co-authored-by: DarkLight1337 <tlleungac@connect.ust.hk>

91eea723

[Frontend][4/n] Improve pooling entrypoints | pooling. (#39153) · 66c079ae
wang.yuqi authored Apr 09, 2026
```
Signed-off-by: wang.yuqi <yuqi.wang@daocloud.io>
```
66c079ae
[XPU] Skip VLLM_BATCH_INVARIANT for XPU in EAGLE DP test (#39164) · e80e6339
sihao_li authored Apr 09, 2026
```
Signed-off-by: sihao.li <sihao.li@intel.com>
Co-authored-by: Kunshang Ji <kunshang.ji@intel.com>
```
e80e6339
[PD][HeteroArch]Fix accuracy issue with CPU_ATTN as Decoder and Flash_ATTN as prefiller (#38935) · ef5a2268
Chendi.Xue authored Apr 08, 2026
```
Signed-off-by: Chendi Xue <chendi.xue@intel.com>
```
ef5a2268

Fix Responses JSON schema alias serialization (#38519) · 2a49284c

noobHappylife authored Apr 09, 2026


Signed-off-by: noobhappylife <aratar1991@hotmail.com>
Co-authored-by: OpenAI Codex <codex@openai.com>

2a49284c

[Model] Update ColModernVBERT to support latest HF checkpoint (#39307) · d37b3787
Ilya Boytsov authored Apr 09, 2026
```
Signed-off-by: Ilya Boytsov <ilyaboytsov1805@gmail.com>
```
d37b3787

[UX] Integrate DeepGEMM into vLLM wheel via CMake (#37980) · eb4205fe

Michael Goin authored Apr 09, 2026


Signed-off-by: mgoin <mgoin64@gmail.com>
Signed-off-by: Michael Goin <mgoin64@gmail.com>
Co-authored-by: Claude <noreply@anthropic.com>

eb4205fe

[W8A8 Block Linear Refactor][2/N] Remove W8A8Fp8BlockLinearOp and adopt Fp8... · 2e9034c9

Maral authored Apr 09, 2026


[W8A8 Block Linear Refactor][2/N] Remove W8A8Fp8BlockLinearOp and adopt Fp8 block linear kernel selections. (#33892)
Signed-off-by: maral <maralbahari.98@gmail.com>
Signed-off-by: Maral <maralbahari.98@gmail.com>

2e9034c9

[Bugfix] FlashInfer MXINT4 MoE crashes, missing do_finalize (#39315) · 8332078c

Benjamin Chislett authored Apr 08, 2026

Signed-off-by: Benjamin Chislett <bchislett@nvidia.com>
Signed-off-by: Benjamin Chislett <chislett.ben@gmail.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>

8332078c

08 Apr, 2026 18 commits

[CI Bug] Fix pre-commit issue in main (#39347) · 3352bf8b
Wentao Ye authored Apr 08, 2026
```
Signed-off-by: yewentao256 <zhyanwentao@126.com>
```
3352bf8b

[BugFix] --max-model-len=-1 causes over-limit requests to hang and starve the... · 7c94ae16

triangleXIV authored Apr 09, 2026


[BugFix] --max-model-len=-1 causes over-limit requests to hang and starve the entire service (#39102)
Signed-off-by: triangle14 <y1019026570@gmail.com>
Signed-off-by: mgoin <mgoin64@gmail.com>
Co-authored-by: mgoin <mgoin64@gmail.com>

7c94ae16

`tests/v1/e2e/spec_decode`: assert async scheduling is used (#39206) · ad05edfb

Rishi Puri authored Apr 08, 2026


Signed-off-by: Rishi Puri <riship@nvidia.com>
Signed-off-by: Rishi Puri <puririshi98@berkeley.edu>
Signed-off-by: sfeng33 <4florafeng@gmail.com>
Co-authored-by: Benjamin Chislett <chislett.ben@gmail.com>
Co-authored-by: Flora Feng <4florafeng@gmail.com>

ad05edfb

[Feature] Batch invariant nvfp4 linear support (#39322) · 20181372
Wentao Ye authored Apr 08, 2026
```
Signed-off-by: yewentao256 <zhyanwentao@126.com>
```
20181372

[MoE] Move DEEP_GEMM into experts/ subdirectory (#39005) · a776a48b

Jackmin801 authored Apr 08, 2026


Signed-off-by: Jackmin801 <ongjackm@gmail.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>

a776a48b

[Tool] `adjust_request` to reasoning parser, and Gemma4 fixes (#39027) · 8477fe42

Ben Browning authored Apr 08, 2026


Signed-off-by: Ben Browning <bbrownin@redhat.com>
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
Co-authored-by: Cursor <cursoragent@cursor.com>

8477fe42

[Perf][Kernel] Persistent TopK scheduler: unified CUDAGraph-safe kernel with... · b55d830e

Roberto L. Castro authored Apr 08, 2026


[Perf][Kernel] Persistent TopK scheduler: unified CUDAGraph-safe kernel with dynamic per-row dispatch - DeepSeek-V3.2 DSA decode (#37421)
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>
Co-authored-by: Lucas Wilkinson <LucasWilkinson@users.noreply.github.com>

b55d830e

[Feature] NUMA binding support for GPU workers (#38635) · 75e01a39

Shengqi Chen authored Apr 09, 2026


Signed-off-by: Shengqi Chen <harry-chen@outlook.com>
Co-authored-by: Jason Li <jasonlizhengjian@gmail.com>
Co-authored-by: Roger Wang <hey@rogerw.io>

75e01a39

[kv_offload+HMA][5/N]: Track group block hashes and block IDs (#37109) · 512c5eb4
Or Ozeri authored Apr 08, 2026
```
Signed-off-by: Or Ozeri <oro@il.ibm.com>
```
512c5eb4
[Bugfix] Fix Gemma4 streaming tool call corruption for split boolean/number values (#39114) · 13151a4d
Flora Feng authored Apr 08, 2026
```
Signed-off-by: sfeng33 <4florafeng@gmail.com>
```
13151a4d
[ROCm] Enable fused_silu_mul_block_quant on ROCm (#38817) · 56c976c1
Gregory Shtrasberg authored Apr 08, 2026
```
Signed-off-by: Gregory Shtrasberg <Gregory.Shtrasberg@amd.com>
```
56c976c1
[Bugfix] Fix V1 logprobs empty strings for multi-byte UTF-8 tokens when logprobs > 0 (#34875) · 8904fc4d
haosdent authored Apr 08, 2026
```
Signed-off-by: haosdent <haosdent@gmail.com>
```
8904fc4d
[CI Failure] pin nomic-embed-text-v1 revision (#39292) · 4e2ab186
wang.yuqi authored Apr 08, 2026
```
Signed-off-by: wang.yuqi <yuqi.wang@daocloud.io>
```
4e2ab186
[Bugfix][Frontend] Fix Gemma4 streaming HTML duplication after tool calls (#38909) · d734445f
yoke authored Apr 08, 2026
```
Signed-off-by: yoke233 <yoke2012@gmail.com>
```
d734445f
[Parser] Migrate response api streaming to unified parser (#38755) · 927975ea
Flora Feng authored Apr 07, 2026
```
Signed-off-by: sfeng33 <4florafeng@gmail.com>
Signed-off-by: Andrew Xia <axia@meta.com>
```
927975ea
[Bugfix] Fix Qwen3 tool parser for Responses API tools (#38848) · 9ea7d670
Flora Feng authored Apr 07, 2026
```
Signed-off-by: sfeng33 <4florafeng@gmail.com>
```
9ea7d670
[release 2.11] Update to torch 2.11 (#34644) · 2111997f
Andrey Talman authored Apr 07, 2026

2111997f
[Model Runner V2] Fuse probabilistic rejection sample kernels (#38496) · 5daf6227
Giancarlo Delfin authored Apr 07, 2026
```
Signed-off-by: Giancarlo Delfin <gdelfin@inferact.ai>
```
5daf6227

07 Apr, 2026 1 commit
- [Attention][V0 Deprecation] Deprecate accept output buffer (#39125) · 70406eb1
  Lucas Wilkinson authored Apr 07, 2026
```
Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
```
  70406eb1