upgrade to vllm 0.11.2 #4400

wangxiyuan · 2025-11-24T09:34:05Z

Bump vLLM version to v0.11.2

What's broken and changed by vLLM:

structured_output is broken by [Core] Async scheduling + structured outputs compatibility vllm#26866
get_mrope_input_positions is broken by [Model] Pass mm_features directly into get_mrope_input_positions vllm#28399
graph mode is broken by Avoid bytecode hook and simplify TorchCompileWrapperWithCustomDipatch vllm#25110 we'll upgrade torch to 2.8 to fix the problem later
embedding is broken by Rename clashing method names for vLLM model protocol vllm#27583
get_attn_backend_cls and attention backend is broken are broken by [CI Failure] Fix backend selection for encoder-only models vllm#28534
spec decode is broken by [Redo] #26368 vllm#28771
sp feature is broken by [compile] Enable sequence parallelism matching w/o custom ops enabled vllm#27126
mtp is broken by [AsyncScheduling] Don't schedule past request max_tokens vllm#27922
lora is broken by [Bugfix][LoRA][Spec Decode] Support LoRA with speculative decoding vllm#21068
execute_model is broken by [Core] Async scheduling + structured outputs compatibility vllm#26866
VLLM_DISABLE_SHARED_EXPERTS_STREAM env is broken by [Bug] Fix env string "0" same to True vllm#28159
kv cahe is broken by [Hybrid] Pass kernel block size to builders vllm#27753
dp is broken by Avoid bytecode hook and simplify TorchCompileWrapperWithCustomDipatch vllm#25110

What's broken and changed by ourself:

qwen vl is broken by [Model][MM] Extract conv layer as CustomOp vllm#28455 We'll remove model files in the future to avoid this kind of error
Engine core is broken by [V1] Support MP Executor for multi node distributed inference vllm#23691 We'll remove the patch file in the future.
Ascend scheduler is broken by [Misc] Make SchedulerConfig.max_model_len init-only vllm#28733 We'll remove ascend scheudler later.
qwen3-next is broken by [PERF] Decouple projections from GDN custom op. Attempt 2 vllm#28083 We'll remove model files in the future to avoid this kind of error
qwen vl is broken by [Bugfix][Qwen][Multimodal] Move Qwen2_5_vl sdpa to custom op and reenable compile vllm#27764. We'll remove model files in the future

Known issue:

ray doesn't work
the accuracy of qwen3-next is not correct
qwen3-vl is broken
prefix cache+ ascend scheduler + deepseek v2 lite is broken.

Co-authored-by: MengqingCao [email protected]
Co-authored-by: hfadzxy [email protected]
Co-authored-by: leo-pony [email protected]
Co-authored-by: 22dimensions [email protected]
Co-authored-by: shen-shanshan [email protected]

vLLM version: v0.11.0
vLLM main: vllm-project/vllm@2918c1b

github-actions · 2025-11-24T09:35:31Z

👋 Hi! Thank you for contributing to the vLLM Ascend project. The following points will speed up your PR merge:‌‌

A PR should do only one thing, smaller PRs enable faster reviews.
Every PR should include unit tests and end-to-end tests ‌to ensure it works and is not broken by other future PRs.
Write the commit message by fulfilling the PR description to help reviewer and future developers understand.

If CI fails, you can run linting and testing checks locally according Contributing and Testing.

gemini-code-assist

Code Review

This pull request upgrades the vLLM dependency to version v0.11.2 and adapts the codebase to the corresponding upstream changes. The modifications are extensive and touch upon various components, including Dockerfiles, documentation, tests, and core implementation files.

Key changes include:

Updating the VLLM_TAG in all Dockerfiles to v0.11.2 and optimizing the git clone process.
A significant architectural refactoring in the model runner, which splits the model execution logic into two distinct steps: execute_model for the forward pass and a new sample_tokens method for token sampling and applying grammar constraints. This change is consistently propagated through the worker and model runner implementations.
Adapting to API changes in SchedulerOutput by removing grammar-related fields, which are now handled in the new sample_tokens step.
Updating the custom attention backend registration to align with vLLM's new decorator-based system.
Refactoring model implementations such as Qwen2.5-VL and Qwen3-Next to align with upstream method renames and simplifications, including the removal of custom workarounds that are no longer necessary.
Enhancements to the MultiprocExecutor to better support multi-node distributed execution.

The changes are well-integrated and appear to correctly adapt the project to the new vLLM version. I have not identified any critical or high-severity issues in this pull request.

zhangxinyuehfad · 2025-11-24T10:55:25Z

@leo-pony
Multi-Node-Ray test failed:
log:

(EngineCore_DP0 pid=300679) (RayWorkerWrapper pid=300872) INFO 11-24 08:50:32 [__init__.py:106] Registered model loader `<class 'vllm_ascend.model_loader.netloader.netloader.ModelNetLoaderElastic'>` with load format `netloader`
(EngineCore_DP0 pid=300679) (RayWorkerWrapper pid=300872) WARNING 11-24 08:50:33 [worker_base.py:301] Missing `shared_worker_lock` argument from executor. This argument is needed for mm_processor_cache_type='shm'.
(EngineCore_DP0 pid=300679) (RayWorkerWrapper pid=300872) INFO 11-24 08:50:33 [utils.py:973] FLASHCOMM2 not enable.
(EngineCore_DP0 pid=300679) ERROR 11-24 08:50:34 [core.py:842] EngineCore failed to start.
(EngineCore_DP0 pid=300679) ERROR 11-24 08:50:34 [core.py:842] Traceback (most recent call last):
(EngineCore_DP0 pid=300679) ERROR 11-24 08:50:34 [core.py:842]   File "/vllm-workspace/vllm/vllm/v1/engine/core.py", line 833, in run_engine_core
(EngineCore_DP0 pid=300679) ERROR 11-24 08:50:34 [core.py:842]     engine_core = EngineCoreProc(*args, **kwargs)
(EngineCore_DP0 pid=300679) ERROR 11-24 08:50:34 [core.py:842]                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(EngineCore_DP0 pid=300679) ERROR 11-24 08:50:34 [core.py:842]   File "/vllm-workspace/vllm/vllm/v1/engine/core.py", line 606, in __init__
(EngineCore_DP0 pid=300679) ERROR 11-24 08:50:34 [core.py:842]     super().__init__(
(EngineCore_DP0 pid=300679) ERROR 11-24 08:50:34 [core.py:842]   File "/vllm-workspace/vllm/vllm/v1/engine/core.py", line 102, in __init__
(EngineCore_DP0 pid=300679) ERROR 11-24 08:50:34 [core.py:842]     self.model_executor = executor_class(vllm_config)
(EngineCore_DP0 pid=300679) ERROR 11-24 08:50:34 [core.py:842]                           ^^^^^^^^^^^^^^^^^^^^^^^^^^^
(EngineCore_DP0 pid=300679) ERROR 11-24 08:50:34 [core.py:842]   File "/vllm-workspace/vllm/vllm/v1/executor/abstract.py", line 101, in __init__
(EngineCore_DP0 pid=300679) ERROR 11-24 08:50:34 [core.py:842]     self._init_executor()
(EngineCore_DP0 pid=300679) ERROR 11-24 08:50:34 [core.py:842]   File "/vllm-workspace/vllm/vllm/v1/executor/ray_executor.py", line 97, in _init_executor
(EngineCore_DP0 pid=300679) ERROR 11-24 08:50:34 [core.py:842]     self._init_workers_ray(placement_group)
(EngineCore_DP0 pid=300679) ERROR 11-24 08:50:34 [core.py:842]   File "/vllm-workspace/vllm/vllm/v1/executor/ray_executor.py", line 370, in _init_workers_ray
(EngineCore_DP0 pid=300679) ERROR 11-24 08:50:34 [core.py:842]     self.collective_rpc("init_device")
(EngineCore_DP0 pid=300679) ERROR 11-24 08:50:34 [core.py:842]   File "/vllm-workspace/vllm/vllm/v1/executor/ray_executor.py", line 493, in collective_rpc
(EngineCore_DP0 pid=300679) ERROR 11-24 08:50:34 [core.py:842]     return ray.get(ray_worker_outputs, timeout=timeout)
(EngineCore_DP0 pid=300679) ERROR 11-24 08:50:34 [core.py:842]            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(EngineCore_DP0 pid=300679) ERROR 11-24 08:50:34 [core.py:842]   File "/usr/local/python3.11.13/lib/python3.11/site-packages/ray/_private/auto_init_hook.py", line 22, in auto_init_wrapper
(EngineCore_DP0 pid=300679) ERROR 11-24 08:50:34 [core.py:842]     return fn(*args, **kwargs)
(EngineCore_DP0 pid=300679) ERROR 11-24 08:50:34 [core.py:842]            ^^^^^^^^^^^^^^^^^^^
(EngineCore_DP0 pid=300679) ERROR 11-24 08:50:34 [core.py:842]   File "/usr/local/python3.11.13/lib/python3.11/site-packages/ray/_private/client_mode_hook.py", line 104, in wrapper
(EngineCore_DP0 pid=300679) ERROR 11-24 08:50:34 [core.py:842]     return func(*args, **kwargs)
(EngineCore_DP0 pid=300679) ERROR 11-24 08:50:34 [core.py:842]            ^^^^^^^^^^^^^^^^^^^^^
(EngineCore_DP0 pid=300679) ERROR 11-24 08:50:34 [core.py:842]   File "/usr/local/python3.11.13/lib/python3.11/site-packages/ray/_private/worker.py", line 2858, in get
(EngineCore_DP0 pid=300679) ERROR 11-24 08:50:34 [core.py:842]     values, debugger_breakpoint = worker.get_objects(object_refs, timeout=timeout)
(EngineCore_DP0 pid=300679) ERROR 11-24 08:50:34 [core.py:842]                                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(EngineCore_DP0 pid=300679) ERROR 11-24 08:50:34 [core.py:842]   File "/usr/local/python3.11.13/lib/python3.11/site-packages/ray/_private/worker.py", line 958, in get_objects
(EngineCore_DP0 pid=300679) ERROR 11-24 08:50:34 [core.py:842]     raise value.as_instanceof_cause()
(EngineCore_DP0 pid=300679) ERROR 11-24 08:50:34 [core.py:842] ray.exceptions.RayTaskError(AssertionError): ray::RayWorkerWrapper.execute_method() (pid=300878, ip=172.22.0.188, actor_id=ccad69f02f06cafa8981145201000000, repr=<vllm.v1.executor.ray_utils.RayWorkerWrapper object at 0xffcfbc328810>)
(EngineCore_DP0 pid=300679) ERROR 11-24 08:50:34 [core.py:842]            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(EngineCore_DP0 pid=300679) ERROR 11-24 08:50:34 [core.py:842]            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(EngineCore_DP0 pid=300679) ERROR 11-24 08:50:34 [core.py:842]   File "/vllm-workspace/vllm/vllm/v1/worker/worker_base.py", line 343, in execute_method
(EngineCore_DP0 pid=300679) ERROR 11-24 08:50:34 [core.py:842]     raise e
(EngineCore_DP0 pid=300679) ERROR 11-24 08:50:34 [core.py:842]   File "/vllm-workspace/vllm/vllm/v1/worker/worker_base.py", line 332, in execute_method
(EngineCore_DP0 pid=300679) ERROR 11-24 08:50:34 [core.py:842]     return run_method(self, method, args, kwargs)
(EngineCore_DP0 pid=300679) ERROR 11-24 08:50:34 [core.py:842]            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(EngineCore_DP0 pid=300679) ERROR 11-24 08:50:34 [core.py:842]   File "/vllm-workspace/vllm/vllm/v1/serial_utils.py", line 479, in run_method
(EngineCore_DP0 pid=300679) ERROR 11-24 08:50:34 [core.py:842]     return func(*args, **kwargs)
(EngineCore_DP0 pid=300679) ERROR 11-24 08:50:34 [core.py:842]            ^^^^^^^^^^^^^^^^^^^^^
(EngineCore_DP0 pid=300679) ERROR 11-24 08:50:34 [core.py:842]            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(EngineCore_DP0 pid=300679) ERROR 11-24 08:50:34 [core.py:842]   File "/vllm-workspace/vllm/vllm/v1/worker/worker_base.py", line 324, in init_device
(EngineCore_DP0 pid=300679) ERROR 11-24 08:50:34 [core.py:842]     self.worker.init_device()  # type: ignore
(EngineCore_DP0 pid=300679) ERROR 11-24 08:50:34 [core.py:842]     ^^^^^^^^^^^^^^^^^^^^^^^^^
(EngineCore_DP0 pid=300679) ERROR 11-24 08:50:34 [core.py:842]   File "/vllm-workspace/vllm-ascend/vllm_ascend/worker/worker_v1.py", line 236, in init_device
(EngineCore_DP0 pid=300679) ERROR 11-24 08:50:34 [core.py:842]     self.device = self._init_device()
(EngineCore_DP0 pid=300679) ERROR 11-24 08:50:34 [core.py:842]                   ^^^^^^^^^^^^^^^^^^^
(EngineCore_DP0 pid=300679) ERROR 11-24 08:50:34 [core.py:842]   File "/vllm-workspace/vllm-ascend/vllm_ascend/worker/worker_v1.py", line 220, in _init_device
(EngineCore_DP0 pid=300679) ERROR 11-24 08:50:34 [core.py:842]     assert self.parallel_config.local_world_size <= visible_device_count, (
(EngineCore_DP0 pid=300679) ERROR 11-24 08:50:34 [core.py:842]            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(EngineCore_DP0 pid=300679) ERROR 11-24 08:50:34 [core.py:842] AssertionError: local_world_size (32) must be less than or equal to the number of visible devices (16).

tests/e2e/multicard/test_offline_inference_distributed.py

vllm_ascend/attention/attention_v1.py

vllm_ascend/patch/platform/patch_multiproc_executor.py

wangxiyuan · 2025-11-24T12:19:32Z

vllm_ascend/worker/worker_v1.py

+
+        visible_device_count = (torch.npu.device_count()
+                                if torch.npu.is_available() else 0)
+        assert self.parallel_config.local_world_size <= visible_device_count, (


github-actions · 2025-11-25T12:19:55Z

This pull request has conflicts, please resolve those before we can evaluate the pull request.

wangxiyuan · 2025-11-26T03:02:57Z

e2e passed here: https://github.com/vllm-project/vllm-ascend/actions/runs/19670970374/job/56399659050?pr=4400

github-actions · 2025-11-26T03:09:05Z

This pull request has conflicts, please resolve those before we can evaluate the pull request.

Signed-off-by: wangxiyuan <[email protected]>

Signed-off-by: MengqingCao <[email protected]>

Signed-off-by: hfadzxy <[email protected]>

Signed-off-by: leo-pony <[email protected]>

…error Signed-off-by: leo-pony <[email protected]>

Signed-off-by: MengqingCao <[email protected]>

Signed-off-by: leo-pony <[email protected]>

Signed-off-by: wangxiyuan <[email protected]>

wangxiyuan · 2025-11-26T03:19:59Z

qwen3-vl broken: https://github.com/vllm-project/vllm-ascend/actions/runs/19670970387/job/56399697329?pr=4400

Signed-off-by: hfadzxy <[email protected]>

MengqingCao

Let's address the known issues in the follow-up prs

Bump vLLM version to v0.11.2 What's broken and changed by vLLM: 1. structured_output is broken by vllm-project/vllm#26866 2. get_mrope_input_positions is broken by vllm-project/vllm#28399 3. graph mode is broken by vllm-project/vllm#25110 we'll upgrade torch to 2.8 to fix the problem later 4. embedding is broken by vllm-project/vllm#27583 5. `get_attn_backend_cls` and attention backend is broken are broken by vllm-project/vllm#28534 6. spec decode is broken by vllm-project/vllm#28771 7. sp feature is broken by vllm-project/vllm#27126 8. mtp is broken by vllm-project/vllm#27922 9. lora is broken by vllm-project/vllm#21068 10. execute_model is broken by vllm-project/vllm#26866 11. `VLLM_DISABLE_SHARED_EXPERTS_STREAM` env is broken by vllm-project/vllm#28159 12. kv cahe is broken by vllm-project/vllm#27753 13. dp is broken by vllm-project/vllm#25110 What's broken and changed by ourself: 1. qwen vl is broken by vllm-project/vllm#28455 We'll remove model files in the future to avoid this kind of error 2. Engine core is broken by vllm-project/vllm#23691 We'll remove the patch file in the future. 3. Ascend scheduler is broken by vllm-project/vllm#28733 We'll remove ascend scheudler later. 4. qwen3-next is broken by vllm-project/vllm#28083 We'll remove model files in the future to avoid this kind of error 5. qwen vl is broken by vllm-project/vllm#27764. We'll remove model files in the future Known issue: 1. ray doesn't work 2. the accuracy of qwen3-next is not correct 3. qwen3-vl is broken 4. prefix cache+ ascend scheduler + deepseek v2 lite is broken. Co-authored-by: MengqingCao <[email protected]> Co-authored-by: hfadzxy <[email protected]> Co-authored-by: leo-pony <[email protected]> Co-authored-by: 22dimensions <[email protected]> Co-authored-by: shen-shanshan <[email protected]> - vLLM version: v0.11.2 --------- Signed-off-by: wangxiyuan <[email protected]> Signed-off-by: MengqingCao <[email protected]> Signed-off-by: hfadzxy <[email protected]> Signed-off-by: leo-pony <[email protected]> Co-authored-by: MengqingCao <[email protected]> Co-authored-by: hfadzxy <[email protected]> Co-authored-by: leo-pony <[email protected]> Signed-off-by: Kurumi5210 <[email protected]>

Bump vLLM version to v0.11.2 What's broken and changed by vLLM: 1. structured_output is broken by vllm-project/vllm#26866 2. get_mrope_input_positions is broken by vllm-project/vllm#28399 3. graph mode is broken by vllm-project/vllm#25110 we'll upgrade torch to 2.8 to fix the problem later 4. embedding is broken by vllm-project/vllm#27583 5. `get_attn_backend_cls` and attention backend is broken are broken by vllm-project/vllm#28534 6. spec decode is broken by vllm-project/vllm#28771 7. sp feature is broken by vllm-project/vllm#27126 8. mtp is broken by vllm-project/vllm#27922 9. lora is broken by vllm-project/vllm#21068 10. execute_model is broken by vllm-project/vllm#26866 11. `VLLM_DISABLE_SHARED_EXPERTS_STREAM` env is broken by vllm-project/vllm#28159 12. kv cahe is broken by vllm-project/vllm#27753 13. dp is broken by vllm-project/vllm#25110 What's broken and changed by ourself: 1. qwen vl is broken by vllm-project/vllm#28455 We'll remove model files in the future to avoid this kind of error 2. Engine core is broken by vllm-project/vllm#23691 We'll remove the patch file in the future. 3. Ascend scheduler is broken by vllm-project/vllm#28733 We'll remove ascend scheudler later. 4. qwen3-next is broken by vllm-project/vllm#28083 We'll remove model files in the future to avoid this kind of error 5. qwen vl is broken by vllm-project/vllm#27764. We'll remove model files in the future Known issue: 1. ray doesn't work 2. the accuracy of qwen3-next is not correct 3. qwen3-vl is broken 4. prefix cache+ ascend scheduler + deepseek v2 lite is broken. Co-authored-by: MengqingCao <[email protected]> Co-authored-by: hfadzxy <[email protected]> Co-authored-by: leo-pony <[email protected]> Co-authored-by: 22dimensions <[email protected]> Co-authored-by: shen-shanshan <[email protected]> - vLLM version: v0.11.2 --------- Signed-off-by: wangxiyuan <[email protected]> Signed-off-by: MengqingCao <[email protected]> Signed-off-by: hfadzxy <[email protected]> Signed-off-by: leo-pony <[email protected]> Co-authored-by: MengqingCao <[email protected]> Co-authored-by: hfadzxy <[email protected]> Co-authored-by: leo-pony <[email protected]>

github-actions bot added documentation Improvements or additions to documentation module:tests module:ops module:core labels Nov 24, 2025

gemini-code-assist bot reviewed Nov 24, 2025

View reviewed changes

wangxiyuan mentioned this pull request Nov 24, 2025

Upgrade vLLM to v0.11.2 #4368

Closed

wangxiyuan changed the title ~~upgrade to vllm new commit~~ upgrade to vllm 0.11.2 Nov 24, 2025

This comment was marked as resolved.

Sign in to view

wangxiyuan force-pushed the 4142-rebase branch from 3c9d947 to eeed2ce Compare November 24, 2025 11:12

wangxiyuan commented Nov 24, 2025

View reviewed changes

wangxiyuan added ready read for review ready-for-test start test by label for PR labels Nov 24, 2025

wangxiyuan force-pushed the 4142-rebase branch from 4ab8bf0 to efe41e8 Compare November 25, 2025 08:55

github-actions bot added the merge-conflicts label Nov 25, 2025

leo-pony force-pushed the 4142-rebase branch from 57a93eb to 588ed71 Compare November 25, 2025 13:06

github-actions bot removed the merge-conflicts label Nov 25, 2025

github-actions bot added the merge-conflicts label Nov 26, 2025

wangxiyuan and others added 8 commits November 26, 2025 11:14

upgrade to vllm new commit

1d26968

Signed-off-by: wangxiyuan <[email protected]>

fix torchair model error

fe9c17c

Signed-off-by: wangxiyuan <[email protected]>

fix patch_multiproc_executor

799cbd5

Signed-off-by: wangxiyuan <[email protected]>

init splitting_ops to []

246ffc1

Signed-off-by: wangxiyuan <[email protected]>

rebase to main

39e8d84

Signed-off-by: wangxiyuan <[email protected]>

fix AscendSharedFusedMoE missing use_dp_chunking

91e45f4

Signed-off-by: MengqingCao <[email protected]>

fix ut test

377278a

Signed-off-by: hfadzxy <[email protected]>

Fix ngram shape wrong problem

3bc01c9

Signed-off-by: leo-pony <[email protected]>

leo-pony and others added 4 commits November 26, 2025 11:14

fix the pin_memory torch alloc not aligned to unaligned tcache chunk …

d23ee7b

…error Signed-off-by: leo-pony <[email protected]>

fix _cudagraph_support introduced by #28479

93577d4

Signed-off-by: MengqingCao <[email protected]>

Fix the rebase to upstream/main wrong import and format

6e24158

Signed-off-by: leo-pony <[email protected]>

fix lint and qwen3-next

8fcc2c9

Signed-off-by: wangxiyuan <[email protected]>

wangxiyuan force-pushed the 4142-rebase branch from 54d0895 to 8fcc2c9 Compare November 26, 2025 03:15

github-actions bot removed the merge-conflicts label Nov 26, 2025

fix ut test

70ea881

Signed-off-by: hfadzxy <[email protected]>

MengqingCao approved these changes Nov 26, 2025

View reviewed changes

wangxiyuan merged commit bc69d7c into vllm-project:main Nov 26, 2025
29 of 40 checks passed

leo-pony mentioned this pull request Nov 26, 2025

[Bug]: Ray start failed with multi-node case #4456

Open

MengqingCao mentioned this pull request Nov 28, 2025

[FixCI] Enable chunked prefill for auto-prefix-caching test #4551

Closed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

upgrade to vllm 0.11.2 #4400

upgrade to vllm 0.11.2 #4400

wangxiyuan commented Nov 24, 2025 •

edited by github-actions bot

Loading

Uh oh!

github-actions bot commented Nov 24, 2025

Uh oh!

gemini-code-assist bot left a comment

Uh oh!

zhangxinyuehfad commented Nov 24, 2025

Uh oh!

This comment was marked as resolved.

Uh oh!

Uh oh!

Uh oh!

wangxiyuan Nov 24, 2025

Uh oh!

github-actions bot commented Nov 25, 2025

Uh oh!

wangxiyuan commented Nov 26, 2025

Uh oh!

github-actions bot commented Nov 26, 2025

Uh oh!

wangxiyuan commented Nov 26, 2025

Uh oh!

MengqingCao left a comment •

edited

Loading

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

upgrade to vllm 0.11.2 #4400

upgrade to vllm 0.11.2 #4400

Conversation

wangxiyuan commented Nov 24, 2025 • edited by github-actions bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

github-actions bot commented Nov 24, 2025

Uh oh!

gemini-code-assist bot left a comment

Choose a reason for hiding this comment

Code Review

Uh oh!

zhangxinyuehfad commented Nov 24, 2025

Uh oh!

This comment was marked as resolved.

Uh oh!

Uh oh!

Uh oh!

wangxiyuan Nov 24, 2025

Choose a reason for hiding this comment

Uh oh!

github-actions bot commented Nov 25, 2025

Uh oh!

wangxiyuan commented Nov 26, 2025

Uh oh!

github-actions bot commented Nov 26, 2025

Uh oh!

wangxiyuan commented Nov 26, 2025

Uh oh!

MengqingCao left a comment • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

wangxiyuan commented Nov 24, 2025 •

edited by github-actions bot

Loading

MengqingCao left a comment •

edited

Loading