vllm-project/vllm

Pull Requests Commits

Synchronize TYPE_CHECKING section with environment_variables dictionary in envs.py

Copilot committed 179 days ago

c6efc2af

Copilot committed 179 days ago

d95d5544

[Misc] Add release note draft to PR template (#23598)

simon-mo committed 179 days ago

Verified 2a97ffc3

[Misc] Simplify FlashInfer attention metadata (#23585)

WoosukKwon committed 180 days ago

Verified efc88cf6

[Docs] Update Documentation of Cohere Command-A Models (#23584)

Terrencezzj committed 180 days ago

Verified 7b6a8372

[TPU][Bugfix] Fixes prompt_token_ids error in tpu tests. (#23574)

patemotter committed 180 days ago

Verified c34c82b7

[XPU] Delay BF16 check to worker init for spawn compatibility (#22979)

chaojun-zhang committed 180 days ago

Verified 8a044754

[Bugfix][V1][P/D]Fix the issue where repeated requests for the same input produce abnormal outputs for P2pNcclConnector (#23403)

Abatom committed 180 days ago

Verified 9188ae7c

[Kernel] Add fused grouped_topk kernel for MoE (#23274)

xyang16 committed 180 days ago

Verified 8a3cd90a

[test][RL] Add sleep level 2 test and fix reload with sleep mode (#23521)

22quinn committed 180 days ago

Verified 2a167b2e

[Refactor] Refactor persistent buffers with CpuGpuBuffer (#23515)

WoosukKwon committed 180 days ago

Verified 0ff902f3

[Bugfix] Fix Qwen3 MoE GPTQ inference (#23490)

Isotr0py committed 180 days ago

Verified a9082a4d

Updates to Flex + VLLm integration (#21416)

drisspg committed 180 days ago

Verified e0329ed4

[Refactor] Pass `tokenizer` explicitly instead of binding to prompt update (#23542)

DarkLight1337 committed 180 days ago

Verified 6879cd80

[Doc] Add caution for API server scale-out (#23550)

DarkLight1337 committed 180 days ago

Verified e269be2b

[Attention] Unify mamba and attention backend selection (#23171)

ayushsatyam146 committed 180 days ago

Verified 5c4b6e66

[misc] add shanghai meetup (#23535)

youkaichao committed 180 days ago

Verified d0a4a3f6

[Bugfix] Allow dynamic number of patches for llava_onevision (#23525)

DarkLight1337 committed 180 days ago

Verified ebafb093

Feature/benchmark/random mm data/images (#23119)

h-brenoskuk committed 180 days ago

Verified 0cb7b065

[Fix] DeepSeek V3.1 tool parser error message (#23492)

skyloevil committed 180 days ago

Verified 2da02dd0

[Core][Multimodal] Track encode cache entries by mm_hash and enable embedding sharing between requests (#22711)

fake0fan committed 180 days ago

Verified d765cf01

[Refactor] Dynamic `target` and `content` for prompt updates (#23411)

DarkLight1337 committed 180 days ago

Verified 712d0f88

[gpt-oss] use reasoning channel for reasoning text in serving_chat (#22920)

yuguo68 committed 180 days ago

Verified 49ab23b3

[Bugfix] Fix Dense module loading for sentence-transformers embedding models (simplified V2) (#23408)

FFFfff1FFFfff committed 180 days ago

Verified c9abb104

Migrate DonutImagePixelInputs to TensorSchema (#23509)

bbeckca committed 180 days ago

Verified 787cdb38

Migrate skyworkr1v inputs to TensorSchema (#23499)

bbeckca committed 180 days ago

Verified a5203d04

Migrate tarsier inputs to TensorSchema (#23500)

bbeckca committed 180 days ago

Verified 99f80944

[Misc] Unified linear print info (#23516)

jeejeelee committed 180 days ago

Verified 170e8ea9

[Bugfix] Fix Qwen2.5-VL quantized model weights loading (#23512)

zifeitong committed 180 days ago

Verified a71e4765

Frontend: Adding LM Format Enforcer support to V1 engine (#22564)

noamgat committed 180 days ago

Verified 39971db3

Older