sync main by tc-mb · Pull Request #5 · tc-mb/vllm

tc-mb · 2026-02-06T11:02:19Z

Purpose

Test Plan

Test Result

Essential Elements of an Effective PR Description Checklist

The purpose of the PR, such as "Fix some issue (link existing issues this PR will resolve)".
The test plan, such as providing test command.
The test results, such as pasting the results comparison before and after, or e2e results
(Optional) The necessary documentation update, such as updating supported_models.md and examples for a new model.
(Optional) Release notes update. If your change is user facing, please update the release notes draft in the Google Doc.

…t#33282) Signed-off-by: Lucas Kabela <lucaskabela@meta.com>

…33372) Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>

Signed-off-by: hujiaxin <524446785@qq.com> Signed-off-by: Emilie1001 <79921183+Emilie1001@users.noreply.github.com> Co-authored-by: Emilie1001 <79921183+Emilie1001@users.noreply.github.com>

Signed-off-by: Tianshu Yu <tianshuyu.formal@gmail.com>

…roject#33396) Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>

Signed-off-by: carlory <baofa.fan@daocloud.io> Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>

…oject#33187) Signed-off-by: Patrick von Platen <patrick.v.platen@gmail.com> Co-authored-by: Nick Hill <nickhill123@gmail.com>

…project#33406) Signed-off-by: juliendenize <julien.denize@mistral.ai>

…oject#33388) Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>

Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>

…project#33323) Signed-off-by: carlory <baofa.fan@daocloud.io> Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>

…#32286) Signed-off-by: Nathan Weinberg <nweinber@redhat.com>

Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>

Signed-off-by: Julien Denize <julien.denize@mistral.ai> Signed-off-by: juliendenize <julien.denize@mistral.ai> Co-authored-by: Michael Goin <mgoin64@gmail.com>

Signed-off-by: frankwang28 <frank.wbb@hotmail.com> Signed-off-by: Frank Wang <41319051+frankwang28@users.noreply.github.com> Co-authored-by: Wentao Ye <44945378+yewentao256@users.noreply.github.com>

…roject#33393) Signed-off-by: Danielle Robinson <dmmaddix@amazon.com> Co-authored-by: Danielle Robinson <dmmaddix@amazon.com>

Signed-off-by: Kyle Sayers <kylesayrs@gmail.com>

…3280) Signed-off-by: mgoin <mgoin64@gmail.com>

Signed-off-by: NickLucche <nlucches@redhat.com>

…33236) Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>

Signed-off-by: Yanan Cao <gmagogsfm@gmail.com>

Signed-off-by: vasiliy <vasiliy@fb.com> Signed-off-by: Michael Goin <mgoin64@gmail.com> Co-authored-by: Michael Goin <mgoin64@gmail.com>

…wen3_next as example models) (vllm-project#33173) Signed-off-by: xuebwang-amd <xuebwang@amd.com>

…ect#32437) Signed-off-by: Pavani Majety <pmajety@nvidia.com>

Signed-off-by: Huy Do <huydhn@gmail.com>

…ainer name (vllm-project#33286) Signed-off-by: Chendi Xue <chendi.xue@intel.com>

Signed-off-by: mgoin <mgoin64@gmail.com>

…t#32831 (vllm-project#33366) Signed-off-by: Gregory Shtrasberg <Gregory.Shtrasberg@amd.com>

Signed-off-by: Alberto Ferrer <albertof@barrahome.org>

…s` in v0.15 (vllm-project#33362) Signed-off-by: yewentao256 <zhyanwentao@126.com>

Signed-off-by: Benjamin Chislett <bchislett@nvidia.com>

Signed-off-by: NickLucche <nlucches@redhat.com>

…llm-project#33795) Signed-off-by: Yoray Zack <yorayz@nvidia.com>

…e constructor (vllm-project#33375) Signed-off-by: Bill Nell <bnell@redhat.com> Co-authored-by: Robert Shaw <114415538+robertgshaw2-redhat@users.noreply.github.com>

Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>

…e log (vllm-project#29714) Signed-off-by: Tsukasa OI <floss_llm@irq.a4lg.com>

…33902) Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>

…onent (vllm-project#33904) Signed-off-by: NickLucche <nlucches@redhat.com>

Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>

Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>

…llm-project#33491) Signed-off-by: Lihao Ran <imlihao.ran@gmail.com> Signed-off-by: mgoin <mgoin64@gmail.com> Co-authored-by: mgoin <mgoin64@gmail.com>

Signed-off-by: Hashem Hashemi <hashem.hashemi@amd.com>

…ect#33832) Signed-off-by: wzhao18 <wzhao18.sz@gmail.com>

…roject#33568)

Signed-off-by: simon-mo <simon.mo@hey.com>

…ject#33511) Signed-off-by: rabi <ramishra@redhat.com>

…-project#31366) Signed-off-by: limingliang <limingliang@stepfun.com> Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk> Co-authored-by: limingliang <limingliang@stepfun.com> Co-authored-by: DarkLight1337 <tlleungac@connect.ust.hk>

Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>

Signed-off-by: Rehan Khan <Rehan.Khan7@ibm.com>

Signed-off-by: Kunshang Ji <kunshang.ji@intel.com>

Signed-off-by: sihao.li <sihao.li@intel.com>

Signed-off-by: Chengcheng Pei <chengchengpei@outlook.com> Signed-off-by: chengchengpei <5881383+chengchengpei@users.noreply.github.com> Co-authored-by: chengchengpei <5881383+chengchengpei@users.noreply.github.com> Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>

vllm-project#32263) Signed-off-by: Gassan <gassan.salama@arm.com>

Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>

Signed-off-by: Xinyu Chen <xinyu1.chen@intel.com>

…t#33982) Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>

Signed-off-by: zhangyue66 <zhangyue66@baidu.com>

Lucaskabela and others added 30 commits January 30, 2026 06:43

[CI] Enable mypy import following for vllm/spec_decode (vllm-projec…

726d897

…t#33282) Signed-off-by: Lucas Kabela <lucaskabela@meta.com>

Explicitly set return_dict for apply_chat_template (vllm-project#…

9432ed8

…33372) Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>

[model] Add support for openPangu7B-VL (vllm-project#32449)

ba45bed

Signed-off-by: hujiaxin <524446785@qq.com> Signed-off-by: Emilie1001 <79921183+Emilie1001@users.noreply.github.com> Co-authored-by: Emilie1001 <79921183+Emilie1001@users.noreply.github.com>

fix: allow LFM2 MoE prefix caching (align) (vllm-project#33376)

f45870b

Signed-off-by: Tianshu Yu <tianshuyu.formal@gmail.com>

[Refactor] Move MM item count validation outside of processor (vllm-p…

c87eac1

…roject#33396) Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>

[Misc] Replace Optional[X] with X | None syntax (vllm-project#33332)

1a7894d

Signed-off-by: carlory <baofa.fan@daocloud.io> Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>

[Realtime API] Adds minimal realtime API based on websockets (vllm-pr…

10152d2

…oject#33187) Signed-off-by: Patrick von Platen <patrick.v.platen@gmail.com> Co-authored-by: Nick Hill <nickhill123@gmail.com>

[BUGFIX] Pixtral cannot be loaded with --limit-mm-per-prompt 0 (vllm-…

8e2ad97

…project#33406) Signed-off-by: juliendenize <julien.denize@mistral.ai>

[Doc] [ROCm] Update Documentation to reflect v0.15.0 release (vllm-pr…

174f167

…oject#33388) Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>

Remove deprecated reasoning_content message field (vllm-project#33402)

c5113f6

Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>

[Misc] Clean up HIDDEN_DEPRECATED_METRICS after metric removal (vllm-…

cf896ae

…project#33323) Signed-off-by: carlory <baofa.fan@daocloud.io> Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>

[Doc] Enhance documentation around CPU container images (vllm-project…

58cb55e

…#32286) Signed-off-by: Nathan Weinberg <nweinber@redhat.com>

Fix test_moe.py for Transformers v5 (vllm-project#33413)

a11bc12

Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>

Improve Mistral format checks. (vllm-project#33253)

ae5b7af

Signed-off-by: Julien Denize <julien.denize@mistral.ai> Signed-off-by: juliendenize <julien.denize@mistral.ai> Co-authored-by: Michael Goin <mgoin64@gmail.com>

Disable Cascade Attention for Batch Invariance (vllm-project#32561)

8f5d512

Signed-off-by: frankwang28 <frank.wbb@hotmail.com> Signed-off-by: Frank Wang <41319051+frankwang28@users.noreply.github.com> Co-authored-by: Wentao Ye <44945378+yewentao256@users.noreply.github.com>

[BugFix][LoRA] TritonExperts is ModularMoEPath for FP8 models (vllm-p…

74898a7

…roject#33393) Signed-off-by: Danielle Robinson <dmmaddix@amazon.com> Co-authored-by: Danielle Robinson <dmmaddix@amazon.com>

[QeRL] Layerwise Reloading (vllm-project#32133)

f857a03

Signed-off-by: Kyle Sayers <kylesayrs@gmail.com>

Support FP8 block quant for CompressedTensorsW8A16Fp8 (vllm-project#3…

fd0e377

…3280) Signed-off-by: mgoin <mgoin64@gmail.com>

[CI] Qwen3-ASR transcriptios tests (vllm-project#33414)

8ece607

Signed-off-by: NickLucche <nlucches@redhat.com>

Fix encoder-decoder model disabling mm processor cache (vllm-project#…

67239c4

…33236) Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>

[Kernel] [Helion] [1/N] Add Helion ConfigManager (vllm-project#32740)

6c1f9e4

Signed-off-by: Yanan Cao <gmagogsfm@gmail.com>

fix QERL attention import path (vllm-project#33432)

3f96fcf

Signed-off-by: vasiliy <vasiliy@fb.com> Signed-off-by: Michael Goin <mgoin64@gmail.com> Co-authored-by: Michael Goin <mgoin64@gmail.com>

[Quantization][ROCm] Fix MoE weight loading to be robust (Qwen3_MoE/Q…

f451b45

…wen3_next as example models) (vllm-project#33173) Signed-off-by: xuebwang-amd <xuebwang@amd.com>

[Hardware][SM100] Add TRTLLM Kernel for INT4 W4A16 Kernel. (vllm-proj…

c3a9752

…ect#32437) Signed-off-by: Pavani Majety <pmajety@nvidia.com>

Indicate compile mode in the benchmark results (vllm-project#32990)

9ca66ec

Signed-off-by: Huy Do <huydhn@gmail.com>

[CI][HPU]accelerate hpu test by skip python re-install and clean cont…

2b46557

…ainer name (vllm-project#33286) Signed-off-by: Chendi Xue <chendi.xue@intel.com>

Refactor NVFP4 Linear utils for ModelOpt and CT (vllm-project#33201)

67ebaff

Signed-off-by: mgoin <mgoin64@gmail.com>

[Bugfix][ROCm] Fixing the skinny gemm dispatch logic from vllm-projec…

31aedfe

…t#32831 (vllm-project#33366) Signed-off-by: Gregory Shtrasberg <Gregory.Shtrasberg@amd.com>

[Bugfix] Fix typo in read_offset variable name (vllm-project#33426)

64a40a7

Signed-off-by: Alberto Ferrer <albertof@barrahome.org>

[Deprecation] Deprecate seed_everything and `scatter_mm_placeholder…

010ec0c

…s` in v0.15 (vllm-project#33362) Signed-off-by: yewentao256 <zhyanwentao@126.com>

benchislett and others added 29 commits February 5, 2026 12:37

[Spec Decode] Unified Parallel Drafting (vllm-project#32887)

af3162d

Signed-off-by: Benjamin Chislett <bchislett@nvidia.com>

[Misc] Add debug logs (vllm-project#33931)

7d8c680

Signed-off-by: NickLucche <nlucches@redhat.com>

[Bugfix] Fix swapped engine_ids in NIXL Llama 4 local attention path (v…

1ee9584

…llm-project#33795) Signed-off-by: Yoray Zack <yorayz@nvidia.com>

[Moe Refactor] Make Inplace Flag for FusedMoEModularKernel part of th…

a57c822

…e constructor (vllm-project#33375) Signed-off-by: Bill Nell <bnell@redhat.com> Co-authored-by: Robert Shaw <114415538+robertgshaw2-redhat@users.noreply.github.com>

[Models] Consolidate Deepseek-OCR2 processor (vllm-project#33909)

87d0d17

Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>

[Bugfix] Suppress non-TTY color output on the process name part of th…

92e7562

…e log (vllm-project#29714) Signed-off-by: Tsukasa OI <floss_llm@irq.a4lg.com>

Fix tokenizer test for renamed attr on Transformers v5 (vllm-project#…

1887acc

…33902) Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>

[Misc] Rename translations to speech_to_text for OAI serving comp…

20f5d18

…onent (vllm-project#33904) Signed-off-by: NickLucche <nlucches@redhat.com>

[Bugfix] Fix DSV3.2 NVFP4 (vllm-project#33932)

4145e50

Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>

[Bugfix] Make MM batching more robust (vllm-project#33817)

116880a

Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>

[Minor] Sort safetensors files to ensure deterministic loading order (v…

42d5d70

…llm-project#33491) Signed-off-by: Lihao Ran <imlihao.ran@gmail.com> Signed-off-by: mgoin <mgoin64@gmail.com> Co-authored-by: mgoin <mgoin64@gmail.com>

Adds padding and perf improvements to wvSplitK_fp8 (vllm-project#33527)

d5c4800

Signed-off-by: Hashem Hashemi <hashem.hashemi@amd.com>

[Bugfix] Fix DeepSeek v3.2 tokenizer outputting None issue (vllm-proj…

91a07ff

…ect#33832) Signed-off-by: wzhao18 <wzhao18.sz@gmail.com>

[Feature] OTEL tracing during loading (vllm-project#31162)

325ab6b

[Perf] Disable clean_logits in deepgemm fp8_mqa_logits kernel (vllm-p…

79028d4

…roject#33568)

[Docs] Add reo analytics (vllm-project#33957)

5819ca8

Signed-off-by: simon-mo <simon.mo@hey.com>

fix(ROCm): Make flash_attn import optional in MLA attention (vllm-pro…

20d7454

…ject#33511) Signed-off-by: rabi <ramishra@redhat.com>

[Misc] Update code for encoder-decoder models (vllm-project#33900)

035a6cb

Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>

[CPU] Add BF16 Kernel type for s390x (vllm-project#33788)

ac04dd3

Signed-off-by: Rehan Khan <Rehan.Khan7@ibm.com>

[XPU][4/N] add mxfp4 moe model support (vllm-project#33679)

7439e4f

Signed-off-by: Kunshang Ji <kunshang.ji@intel.com>

[XPU]Replace pip in docker.xpu with uv pip (vllm-project#31112)

6550815

Signed-off-by: sihao.li <sihao.li@intel.com>

[cpu][performance] CPU Paged Attention NEON BFMMLA BF16 Implementation (

1363e3d

vllm-project#32263) Signed-off-by: Gassan <gassan.salama@arm.com>

Fix main pre-commit (vllm-project#33975)

6d8d34b

Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>

support view_from_cpu_tensor on XPU (vllm-project#33868)

e969a16

Signed-off-by: Xinyu Chen <xinyu1.chen@intel.com>

Consolidate and fix forbidden import pre-commit checks (vllm-projec…

791a94b

…t#33982) Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>

[PaddleOCR-VL] Add BC for transformers 5.0 config (vllm-project#33976)

233b26a

Signed-off-by: zhangyue66 <zhangyue66@baidu.com>

Merge branch 'Support-MiniCPM-o-4.5' into main

512c947

tc-mb merged commit cc786bf into Support-MiniCPM-o-4.5 Feb 6, 2026
2 of 3 checks passed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

sync main#5

sync main#5
tc-mb merged 939 commits intoSupport-MiniCPM-o-4.5from
main

tc-mb commented Feb 6, 2026 •

edited by github-actions bot

Loading

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

20 participants

Comments

Conversation

tc-mb commented Feb 6, 2026 • edited by github-actions bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Purpose

Test Plan

Test Result

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

20 participants

Comments

tc-mb commented Feb 6, 2026 •

edited by github-actions bot

Loading