Skip to content

Flashrl v0.11.0 Upgrade#3

Open
devpatelio wants to merge 1299 commits into
mainfrom
flashrl-v0.11.0
Open

Flashrl v0.11.0 Upgrade#3
devpatelio wants to merge 1299 commits into
mainfrom
flashrl-v0.11.0

Conversation

@devpatelio

@devpatelio devpatelio commented Nov 28, 2025 •
edited
Loading

Copy link
Copy Markdown

We made a branch called flashrl-v0.11.0 (base copy of vllm 0.11.0) and looked to merge our local fork's main branch into our flashrl-v0.11.0 branch.

WoosukKwon and others added 30 commits September 21, 2025 08:52
@WoosukKwon
[V0 Deprecation] Remove async_output_proc, preemption mode, delay fac…
0ff8ebb 
…tor (vllm-project#25334)

Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
@rahul-tuli @claude
feat: Enable engine-level arguments with speculators models (vllm-pro…
c438b29 
…ject#25250)

Signed-off-by: Rahul Tuli <rtuli@redhat.com>
Co-authored-by: Claude <noreply@anthropic.com>
@WoosukKwon
[V0 Deprecation] Remove V0 sampling metadata (vllm-project#25345)
1c3ffdb 
Signed-off-by: Woosuk Kwon <woosuk@thinkingmachines.ai>
@Isotr0py
[Perf] Further optimization for Qwen3-VL fast_pos_embed_interpolate (…
af7dfb0 
…vllm-project#25347)

Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
@WoosukKwon
Remove V0 attention backends (vllm-project#25351)
bc6e542 
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
@KKSK-DON
[Bugfix][V0 Deprecation][CI] use async mock and await for async method (
04d3752 
vllm-project#25325)

Signed-off-by: Yang <lymailforjob@gmail.com>
@debroy-rh
Multimodal - audio tests (vllm-project#25285)
5aeb925 
Signed-off-by: Debolina Roy <debroy@redhat.com>
@ywang96 @yinz-aizip
[Model] Support Dots OCR (vllm-project#24645)
7b57a43 
Signed-off-by: Roger Wang <hey@rogerw.io>
Co-authored-by: yinz-aizip <yinz@aizip.ai>
@david6666666
[Docs] GSM8K Accuracy Evaluation doc update (vllm-project#25360)
793be8d 
Signed-off-by: David Chen <530634352@qq.com>
@david6666666 @BruceW-07 @chaunceyjiang
[Bugfix] Fix hermes tool parser handling of non-string argument types (…
0eecb31 
…vllm-project#22002)

Signed-off-by: wangzi <3220100013@zju.edu.cn>
Signed-off-by: David Chen <530634352@qq.com>
Co-authored-by: wangzi <3220100013@zju.edu.cn>
Co-authored-by: Chauncey <chaunceyjiang@gmail.com>
@DarkLight1337
[V0 Deprecation] Remove V0-only methods in multi-modal registry (vllm…
6d0b827 
…-project#25362)

Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
@DarkLight1337
[V0 Deprecation] Remove MultiModalPlaceholderMap (vllm-project#25366)
f92d952 
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
@eldarkurtic
Enable Eagle3 speculative decoding for GPT-OSS model (vllm-project#25246
21467f9 
)

Signed-off-by: Eldar Kurtic <8884008+eldarkurtic@users.noreply.github.com>
@NickLucche
[TPU][Bugfix][CI] Fix broken tests/build dependency (vllm-project#25255)
a66d131 
Signed-off-by: NickLucche <nlucches@redhat.com>
@NickLucche @yeqcharlotte
[TPU] Deprecate xm.mark_step in favor of `torch_xla.sync (vllm-pr…
4cf71cc 
…oject#25254)

Signed-off-by: NickLucche <nlucches@redhat.com>
Co-authored-by: Ye (Charlotte) Qi <yeq@meta.com>
@yiz-liu
refactor: abstract graph mode support into platform interface (vllm-p…
b6f01bd 
…roject#25161)

Signed-off-by: Yizhou Liu <liu_yizhou@outlook.com>
@DarkLight1337
[Misc] Remove unused encoder-decoder error strings (vllm-project#25374)
417a164 
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
@hmellor
Make pickle import check fast (vllm-project#25379)
64c824c 
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
@hmellor @gemini-code-assist
Make mypy behave like a proper pre-commit hook (vllm-project#25313)
3d2c56b 
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
@Sara-KS
[Kernel] MI-300X triton moe configs (vllm-project#23445)
ac24388 
Signed-off-by: Sara Kokkila Schumacher <saraks@ibm.com>
@Csrayz @ivyilike
[Bugfix] Fix several issues with p2p xPyD in GET type (vllm-project#2…
c10101a 
…3993)

Signed-off-by: Csrayz <jover@cmbchina.com>
Signed-off-by: ivyilike <pww123@cmbchina.com>
Co-authored-by: ivyilike <pww123@cmbchina.com>
@bringlein
[V1][Attention] Split triton_attn in triton-only and rocm specific ba…
175811e 
…ckends (vllm-project#24648)

Signed-off-by: Burkhard Ringlein <ngl@zurich.ibm.com>
@abmfy @gemini-code-assist @tlrmchlsmth
[EPLB] Reduce EPLB Inference Overhead (vllm-project#24573)
06a4133 
Signed-off-by: Bowen Wang <abmfy@icloud.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
Co-authored-by: Tyler Michael Smith <tyler@neuralmagic.com>
@Daisy-Ma-coder @DarkLight1337
[CLI env var] Add VLLM_FLASH_ATTN_MAX_NUM_SPLITS_FOR_CUDA_GRAPH in en…
cfbee3d 
…v variables (vllm-project#25274)

Signed-off-by: qqma <qqma@amazon.com>
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
Co-authored-by: qqma <qqma@amazon.com>
Co-authored-by: Cyrus Leung <tlleungac@connect.ust.hk>
@ElizaWszola
[Compiler] Disable Inductor standalone compile by default (vllm-proje…
1d7f95b 
…ct#25391)

Signed-off-by: ElizaWszola <ewszola@redhat.com>
@mgoin
[CI Failure] Fix fp8 kv cache on <SM90 (vllm-project#25396)
239ef0c 
Signed-off-by: mgoin <mgoin64@gmail.com>
@luccafong @zhuohan123
[DP] support torchrun external launcher with Data Parallelism (vllm-p…
922979b 
…roject#24899)

Signed-off-by: Lu Fang <fanglu@fb.com>
Signed-off-by: Zhuohan Li <zhuohan123@gmail.com>
Co-authored-by: Zhuohan Li <zhuohan123@gmail.com>
@simon-mo
[misc] Remove RFC review hours reference (vllm-project#25416)
8d0ee5a
@ProExpertProg
[torch.compile] Cleanup compilation tests and custom passes, add debu…
d5e0fca 
…g utils, fix DCE bug (vllm-project#23091), fix test (vllm-project#24376), and prep for custom op matching (vllm-project#24604) (vllm-project#24542)

Signed-off-by: Luka Govedič <lgovedic@redhat.com>
Signed-off-by: luka <lgovedic@redhat.com>
Signed-off-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
@orozery
[KV offload][5/N] Add CPUOffloadingSpec (vllm-project#24251)
8db2939 
Signed-off-by: Or Ozeri <oro@il.ibm.com>
mgoin and others added 29 commits September 27, 2025 23:32
@mgoin @simon-mo
[CI] Fix FlashInfer AOT in release docker image (vllm-project#25730)
5aa5811 
Signed-off-by: mgoin <mgoin64@gmail.com>
Signed-off-by: simon-mo <simon.mo@hey.com>
@tlrmchlsmth @simon-mo
[Bugfix][WideEP] Apply TP Attn + EP MoE fix to other models (vllm-pro…
26a7a33 
…ject#24982)

Signed-off-by: Tyler Michael Smith <tlrmchlsmth@gmail.com>
Signed-off-by: simon-mo <simon.mo@hey.com>
@NickLucche @simon-mo
[Bugfix][NIXL] Fix Async Scheduler timeout issue (vllm-project#25808)
b14773b 
Signed-off-by: NickLucche <nlucches@redhat.com>
Signed-off-by: simon-mo <simon.mo@hey.com>
@ywang96 @simon-mo
[MM] Optimize memory profiling for scattered multimodal embeddings (v…
6de3d43 
…llm-project#25810)

Signed-off-by: Roger Wang <hey@rogerw.io>
Signed-off-by: simon-mo <simon.mo@hey.com>
@ywang96 @simon-mo
[Bugfix] Fix Qwen3-VL regression from vllm-project#24982 (vllm-projec…
19e7ab7 
…t#25814)

Signed-off-by: Roger Wang <hey@rogerw.io>
Signed-off-by: simon-mo <simon.mo@hey.com>
@Isotr0py @ywang96 @simon-mo
[VLM] Update Qwen3-VL max_num_video_tokens calculation for configurab…
4c34704 
…le video profiling (vllm-project#25557)

Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
Signed-off-by: Roger Wang <hey@rogerw.io>
Co-authored-by: Roger Wang <hey@rogerw.io>
Signed-off-by: simon-mo <simon.mo@hey.com>
@JJJYmmm @simon-mo
[Bugfix] fix Qwen3VLMoe load when pp > 1 (vllm-project#25838)
09c2cbc 
Signed-off-by: liuye.hj <liuye.hj@alibaba-inc.com>
Co-authored-by: liuye.hj <liuye.hj@alibaba-inc.com>
Signed-off-by: simon-mo <simon.mo@hey.com>
@robertgshaw2-redhat @SageMoore
@russellb @NiuBlibing @Isotr0py @huachenheli @yewentao256 @mgoin @tlrmchlsmth @NickLucche @ywang96 @simon-mo
[P/D] NIXL Updates (vllm-project#25844)
8ce5d31 
Signed-off-by: Sage Moore <sage@neuralmagic.com>
Signed-off-by: simon-mo <simon.mo@hey.com>
Signed-off-by: rentianyue-jk <rentianyue-jk@360shuke.com>
Signed-off-by: Russell Bryant <rbryant@redhat.com>
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
Signed-off-by: Chenheli Hua <huachenheli@outlook.com>
Signed-off-by: mgoin <mgoin64@gmail.com>
Signed-off-by: Tyler Michael Smith <tlrmchlsmth@gmail.com>
Signed-off-by: NickLucche <nlucches@redhat.com>
Signed-off-by: Roger Wang <hey@rogerw.io>
Signed-off-by: Robert Shaw <robshaw@redhat.com>
Co-authored-by: Sage Moore <sage@neuralmagic.com>
Co-authored-by: Russell Bryant <rbryant@redhat.com>
Co-authored-by: rentianyue-jk <rentianyue-jk@360shuke.com>
Co-authored-by: Isotr0py <mozf@mail2.sysu.edu.cn>
Co-authored-by: Chenheli Hua <huachenheli@outlook.com>
Co-authored-by: Wentao Ye <44945378+yewentao256@users.noreply.github.com>
Co-authored-by: Michael Goin <mgoin64@gmail.com>
Co-authored-by: Tyler Michael Smith <tlrmchlsmth@gmail.com>
Co-authored-by: Nicolò Lucchesi <nlucches@redhat.com>
Co-authored-by: Roger Wang <hey@rogerw.io>
Co-authored-by: Robert Shaw <robshaw@redhat.com>
Signed-off-by: simon-mo <simon.mo@hey.com>
@ywang96 @simon-mo
[Bugfix] Fallback ViT attn backend to SDPA for blackwell (vllm-projec…
ab5b645 
…t#25851)

Signed-off-by: Roger Wang <hey@rogerw.io>
Signed-off-by: simon-mo <simon.mo@hey.com>
@yewentao256 @gemini-code-assist @simon-mo
[Bug] Fix Weight Loading for Block FP8 Cutlass SM90 (vllm-project#25909)
9471879 
Signed-off-by: yewentao256 <zhyanwentao@126.com>
Signed-off-by: Wentao Ye <44945378+yewentao256@users.noreply.github.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
Signed-off-by: simon-mo <simon.mo@hey.com>
@LucasWilkinson @simon-mo
[BugFix] Fix DP/EP hang (vllm-project#25906)
03df0fb 
Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
Signed-off-by: simon-mo <simon.mo@hey.com>
@zyongye @heheda12345
@youkaichao @LucasWilkinson @robertgshaw2-redhat @yewentao256 @mgoin @luccafong @NickLucche @IwakuraRein @MatthewBonanni @mxz297 @Barry-Delaney @simon-mo
[New Model] DeepSeek-V3.2 (Rebased to Main) (vllm-project#25896)
b3230e1 
Signed-off-by: Chen Zhang <zhangch99@outlook.com>
Signed-off-by: youkaichao <youkaichao@gmail.com>
Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
Signed-off-by: mgoin <mgoin64@gmail.com>
Signed-off-by: NickLucche <nlucches@redhat.com>
Signed-off-by: Yongye Zhu <zyy1102000@gmail.com>
Signed-off-by: Barry Kang <43644113+Barry-Delaney@users.noreply.github.com>
Signed-off-by: Lucia Fang <fanglu@meta.com>
Co-authored-by: Chen Zhang <zhangch99@outlook.com>
Co-authored-by: youkaichao <youkaichao@gmail.com>
Co-authored-by: Lucas Wilkinson <lwilkins@redhat.com>
Co-authored-by: Robert Shaw <114415538+robertgshaw2-redhat@users.noreply.github.com>
Co-authored-by: Lucas Wilkinson <LucasWilkinson@users.noreply.github.com>
Co-authored-by: yewentao256 <zhyanwentao@126.com>
Co-authored-by: Wentao Ye <44945378+yewentao256@users.noreply.github.com>
Co-authored-by: mgoin <mgoin64@gmail.com>
Co-authored-by: Lucia Fang <116399278+luccafong@users.noreply.github.com>
Co-authored-by: Lucia Fang <fanglu@meta.com>
Co-authored-by: NickLucche <nlucches@redhat.com>
Co-authored-by: Siyuan Fu <siyuanf@nvidia.com>
Co-authored-by: Matthew Bonanni <mbonanni@redhat.com>
Co-authored-by: Xiaozhu Meng <mxz297@gmail.com>
Co-authored-by: Barry Kang <43644113+Barry-Delaney@users.noreply.github.com>
Signed-off-by: simon-mo <simon.mo@hey.com>
@NickLucche @heheda12345 @simon-mo
[NIXL] Add support for MLA caches with different latent dim (vllm-pro…
d0b178c 
…ject#25902)

Signed-off-by: NickLucche <nlucches@redhat.com>
Signed-off-by: Chen Zhang <zhangch99@outlook.com>
Co-authored-by: Chen Zhang <zhangch99@outlook.com>
Signed-off-by: simon-mo <simon.mo@hey.com>
@youkaichao @simon-mo
[bugfix][deepseek] fix flashmla kernel selection (vllm-project#25956)
83f3c9b 
Signed-off-by: youkaichao <youkaichao@gmail.com>
Signed-off-by: simon-mo <simon.mo@hey.com>
@jeejeelee @simon-mo
[Bench] Add DeepSeekV32 to MoE benchmark (vllm-project#25962)
c3dfb0f 
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
Signed-off-by: simon-mo <simon.mo@hey.com>
@zixi-qi @simon-mo
[spec decode] Consolidate speculative decode method name for MTP (vll…
c214d69 
…m-project#25232)

Signed-off-by: zixi-qi <qizixi@meta.com>
@luccafong @simon-mo
[Model] MTP fallback to eager for DeepSeek v32 (vllm-project#25982)
bab9231 
Signed-off-by: Lu Fang <fanglu@fb.com>
Signed-off-by: simon-mo <simon.mo@hey.com>
@ywang96 @simon-mo
[MM] Add text-only mode for Qwen3-VL (vllm-project#26000)
a1825fe 
Signed-off-by: simon-mo <simon.mo@hey.com>
@zhewenl @simon-mo
[Bugfix] Fix __syncwarp on ROCM (vllm-project#25996)
febb688 
Signed-off-by: simon-mo <simon.mo@hey.com>
@LucasWilkinson @simon-mo
[BugFix] Fix default kv-cache-dtype default for DeepseekV3.2 (vllm-pr…
e4beabd 
…oject#25988)

Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
Signed-off-by: simon-mo <simon.mo@hey.com>
@LucasWilkinson @simon-mo
[BugFix][DP/EP] Fix CUTLASS MLA hang under load (vllm-project#26026)
ebce361 
Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
Co-authored-by: Robert Shaw <robshaw@redhat.com>
Co-authored-by: rshaw@neuralmagic.com <rshaw@neuralmagic.com>
Signed-off-by: simon-mo <simon.mo@hey.com>
@LucasWilkinson @simon-mo
[BugFix] ChunkedLocalAttention is currently not CG compatible (vllm-p…
c536881 
…roject#26034)

Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
Signed-off-by: simon-mo <simon.mo@hey.com>
@huydhn @simon-mo
Update base image to 22.04 (jammy) (vllm-project#26065)
05bf0c5 
Signed-off-by: Huy Do <huydhn@gmail.com>
Signed-off-by: simon-mo <simon.mo@hey.com>
@LucasWilkinson @gemini-code-assist
@mgoin @simon-mo
[BugFix] Fix FI accuracy issue when used for MLA prefill (vllm-projec…
6040e0b 
…t#26063)

Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
Signed-off-by: Lucas Wilkinson <LucasWilkinson@users.noreply.github.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
Co-authored-by: mgoin <mgoin64@gmail.com>
Signed-off-by: simon-mo <simon.mo@hey.com>
@huachenheli @simon-mo
[Small] Prevent bypassing media domain restriction via HTTP redirects (…
9d9a2b7 
…vllm-project#26035)

Signed-off-by: Chenheli Hua <huachenheli@outlook.com>
Signed-off-by: simon-mo <simon.mo@hey.com>
@heheda12345 @simon-mo
[Deepseek v3.2] Support indexer prefill chunking (vllm-project#25999)
c75c2e7 
Signed-off-by: Chen Zhang <zhangch99@outlook.com>
Signed-off-by: simon-mo <simon.mo@hey.com>
@mgoin @gemini-code-assist @simon-mo
[Bugfix] Disable cascade attention with FlashInfer (vllm-project#26130)
d100776 
Signed-off-by: mgoin <mgoin64@gmail.com>
Signed-off-by: Michael Goin <mgoin64@gmail.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
Signed-off-by: simon-mo <simon.mo@hey.com>
@tlrmchlsmth @simon-mo
[Build/CI] Revert back to Ubuntu 20.04, install python 3.12 with uv (v…
f71952c 
…llm-project#26103)

Signed-off-by: Tyler Michael Smith <tlrmchlsmth@gmail.com>
Co-authored-by: Simon Mo <simon.mo@hey.com>
Signed-off-by: simon-mo <simon.mo@hey.com>
@wseaton @simon-mo
Update CUDA architecture list in build pipeline for 12.9.1 wheels (vl…
b8b302c 
…lm-project#26592)

Signed-off-by: Will Eaton <wseaton@users.noreply.github.com>
Signed-off-by: simon-mo <simon.mo@hey.com>
@github-actions

github-actions Bot commented Nov 28, 2025

Copy link
Copy Markdown

👋 Hi! Thank you for contributing to the vLLM project.

💬 Join our developer Slack at https://slack.vllm.ai to discuss your PR in #pr-reviews, coordinate on features in #feat- channels, or join special interest groups in #sig- channels.

Just a reminder: PRs would not trigger full CI run by default. Instead, it would only run fastcheck CI which starts running only a small and essential subset of CI tests to quickly catch errors. You can run other CI tests on top of those by going to your fastcheck build on Buildkite UI (linked in the PR checks section) and unblock them. If you do not have permission to unblock, ping simon-mo or khluu to add you in our Buildkite org.

Once the PR is approved and ready to go, your PR reviewer(s) can run CI to test the changes comprehensively before merging.

To run CI, PR reviewers can either: Add ready label to the PR or enable auto-merge.

🚀

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

No reviews

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.