Commits

Branch selector

 main

User selector

All users

Datepicker

All time

Commit History

Commits on Aug 25, 2025

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

  Show description for 2a97ffc
  simon-mo

  authored
  2a97ffc

  Copy full SHA for 2a97ffc

• [Misc] Simplify FlashInfer attention metadata (#23585)

  Show description for efc88cf
  WoosukKwon

  authored
  efc88cf

  Copy full SHA for efc88cf

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

  Show description for 7b6a837
  

  Terrencezzj

  and

  Abatom

  authored
  7b6a837

  Copy full SHA for 7b6a837

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

  Show description for c34c82b
  patemotter

  authored
  c34c82b

  Copy full SHA for c34c82b

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

  Show description for 8a04475
  chaojun-zhang

  authored
  8a04475

  Copy full SHA for 8a04475

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

  Show description for 9188ae7
  Abatom

  authored
  9188ae7

  Copy full SHA for 9188ae7

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

  Show description for 8a3cd90
  

  xyang16

  and

  yewentao256

  authored
  8a3cd90

  Copy full SHA for 8a3cd90

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

  Show description for 2a167b2
  22quinn

  authored
  2a167b2

  Copy full SHA for 2a167b2

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

  Show description for 0ff902f
  

  WoosukKwon

  and

  gemini-code-assist[bot]

  authored
  0ff902f

  Copy full SHA for 0ff902f

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

  Show description for a9082a4
  Isotr0py

  authored
  a9082a4

  Copy full SHA for a9082a4

• Updates to Flex + VLLm integration (#21416)

  Show description for e0329ed
  drisspg

  authored
  e0329ed

  Copy full SHA for e0329ed

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

  Show description for 6879cd8
  DarkLight1337

  authored
  6879cd8

  Copy full SHA for 6879cd8

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

  Show description for e269be2
  DarkLight1337

  authored
  e269be2

  Copy full SHA for e269be2

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

  Show description for 5c4b6e6
  ayushsatyam146

  authored
  5c4b6e6

  Copy full SHA for 5c4b6e6

• [misc] add shanghai meetup (#23535)

  Show description for d0a4a3f
  youkaichao

  authored
  d0a4a3f

  Copy full SHA for d0a4a3f

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

  Show description for ebafb09
  DarkLight1337

  authored
  ebafb09

  Copy full SHA for ebafb09

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

  Show description for 0cb7b06
  h-brenoskuk

  authored
  0cb7b06

  Copy full SHA for 0cb7b06

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

  Show description for 2da02dd
  skyloevil

  authored
  2da02dd

  Copy full SHA for 2da02dd

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

  Show description for d765cf0
  

  3 peopleauthored
  d765cf0

  Copy full SHA for d765cf0

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

  Show description for 712d0f8
  DarkLight1337

  authored
  712d0f8

  Copy full SHA for 712d0f8

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

  Show description for 49ab23b
  yuguo68

  authored
  49ab23b

  Copy full SHA for 49ab23b

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

  Show description for c9abb10
  FFFfff1FFFfff

  authored
  c9abb10

  Copy full SHA for c9abb10

• Migrate DonutImagePixelInputs to TensorSchema (#23509)

  Show description for 787cdb3
  bbeckca

  authored
  787cdb3

  Copy full SHA for 787cdb3

• Migrate skyworkr1v inputs to TensorSchema (#23499)

  Show description for a5203d0
  bbeckca

  authored
  a5203d0

  Copy full SHA for a5203d0

• Migrate tarsier inputs to TensorSchema (#23500)

  Show description for 99f8094
  bbeckca

  authored
  99f8094

  Copy full SHA for 99f8094

• [Misc] Unified linear print info (#23516)

  Show description for 170e8ea
  jeejeelee

  authored
  170e8ea

  Copy full SHA for 170e8ea

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

  Show description for a71e476
  zifeitong

  authored
  a71e476

  Copy full SHA for a71e476

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

  Show description for 39971db
  

  3 peopleauthored
  39971db

  Copy full SHA for 39971db

• [Perf] Add Triton config for DeepSeek V3 FP8 EP32 H200 (#23504)

  Show description for 504d914
  minosfuture

  authored
  504d914

  Copy full SHA for 504d914

• [Doc: ]fix various typos in multiple files (#23487)

  Show description for 47455c4
  

  didier-durand

  and

  hmellor

  authored
  47455c4

  Copy full SHA for 47455c4

Commits on Aug 24, 2025

• fix incompatibililty with non cuda platform for nvfp4 (#23478)

  Show description for c7fc6b1
  

  luccafong

  and

  Lucia (Lu) Fang

  authored
  c7fc6b1

  Copy full SHA for c7fc6b1

• [Misc] Remove unused slot_mapping buffer (#23502)

  Show description for ad78868
  WoosukKwon

  authored
  ad78868

  Copy full SHA for ad78868

• [Model] Enable BLOOM on V1 (#23488)

  Show description for e2db116
  DarkLight1337

  authored
  e2db116

  Copy full SHA for e2db116

• [New Model]Donut model (#23229)

  Show description for 416f059
  princepride

  authored
  416f059

  Copy full SHA for 416f059

• (Misc): add missing test for zero truncation size. (#23457)

  Show description for 5e021b4
  teekenl

  authored
  5e021b4

  Copy full SHA for 5e021b4

Pagination

Previous

Next