Commits

Branch selector

 main

User selector

All users

Datepicker

All time

Commit History

Commits on Feb 22, 2025

• [CI/Build] Fix pre-commit errors (vllm-project#13696)

  DarkLight1337

  authored
  7f6bae5

  Copy full SHA for 7f6bae5

• [Misc] Reduce LoRA-related static variable (vllm-project#13166)

  jeejeelee

  authored
  105b8ce

  Copy full SHA for 105b8ce

• [Metrics] Add --show-hidden-metrics-for-version CLI arg (vllm-project#13295)

  markmc

  authored
  2cb8c15

  Copy full SHA for 2cb8c15

• [V1][Metrics] Support vllm:cache_config_info (vllm-project#13299)

  markmc

  authored
  1cd981d

  Copy full SHA for 1cd981d

• Correction to TP logic for Mamba Mixer 2 when Num Groups not divisible by TP Size (vllm-project#13660)

  fabianlim

  authored
  fca2084

  Copy full SHA for fca2084

• [Bugfix] V1 Memory Profiling: V0 Sampler Integration without Rejection Sampler (vllm-project#13594)

  Show description for da31b53
  

  JenZhao

  and

  ywang96

  authored
  da31b53

  Copy full SHA for da31b53

• [v1] Support allowed_token_ids in v1 Sampler (vllm-project#13210)

  Show description for bb78fb3
  houseroad

  authored
  bb78fb3

  Copy full SHA for bb78fb3

• [Bugfix] Fix benchmark script bug: inaccurate stats for vllm backend when max_model_len < input_len + output_len (vllm-project#13691)

  Show description for 8aca27f
  WangErXiao

  authored
  8aca27f

  Copy full SHA for 8aca27f

• [Misc] Bump compressed-tensors (vllm-project#13619)

  dsikka

  authored
  95c617e

  Copy full SHA for 95c617e

• [Bugfix][Model] OLMo 2: split qkv correctly for GQA and MQA (vllm-project#13687)

  2015aroras

  authored
  9a1f1da

  Copy full SHA for 9a1f1da

• [ROCM] fix native attention function call (vllm-project#13650)

  gongdao123

  authored
  68d630a

  Copy full SHA for 68d630a

• [Misc] Capture and log the time of loading weights (vllm-project#13666)

  waltforme

  authored
  68d535e

  Copy full SHA for 68d535e

• [Bugfix][API Server] Fix invalid usage of 'ge' and 'le' in port valid… (vllm-project#13672)

  WangErXiao

  authored
  c6ed938

  Copy full SHA for c6ed938

• [HTTP Server] Make model param optional in request (vllm-project#13568)

  youngkent

  authored
  0ffdf8c

  Copy full SHA for 0ffdf8c

• docs: Add a note on full CI run in contributing guide (vllm-project#13646)

  terrytangyuan

  authored
  8c0dd3d

  Copy full SHA for 8c0dd3d

• [Misc] Fix yapf linting tools etc not running on pre-commit (vllm-project#13695)

  Show description for ada7c78
  Isotr0py

  authored
  ada7c78

  Copy full SHA for ada7c78

Commits on Feb 21, 2025

• [Attention] MLA with chunked prefill (vllm-project#12639)

  Show description for 288cc6c
  

  4 peopleauthored
  288cc6c

  Copy full SHA for 288cc6c

• fix typo of grafana dashboard, with correct datasource (vllm-project#13668)

  Show description for 900edbf
  johnzheng1975

  authored
  900edbf

  Copy full SHA for 900edbf

• [Bugfix][CPU] Fix cpu all-reduce using native pytorch implementation (vllm-project#13586)

  Isotr0py

  authored
  b2c3fc5

  Copy full SHA for b2c3fc5

• [Kernel]Add streamK for block-quantized CUTLASS kernels (vllm-project#12978)

  Hongbosherlock

  authored
  839b27c

  Copy full SHA for 839b27c

• [ci] Fix metrics test model path (vllm-project#13635)

  khluu

  authored
  34ad27f

  Copy full SHA for 34ad27f

• [FEATURE] Enables /score endpoint for embedding models (vllm-project#12846)

  gmarinho2

  authored
  1c3c975

  Copy full SHA for 1c3c975

• Missing comment explaining VDR variable in GGUF kernels (vllm-project#13290)

  SzymonOzog

  authored
  1cdc886

  Copy full SHA for 1cdc886

• [V1][Sampler] Avoid an operation during temperature application (vllm-project#13587)

  njhill

  authored
  31aa045

  Copy full SHA for 31aa045

• [Bugfix] Add mm_processor_kwargs to chat-related protocols (vllm-project#13644)

  ywang96

  authored
  a30c093

  Copy full SHA for a30c093

• Use pre-commit to update requirements-test.txt (vllm-project#13617)

  hmellor

  authored
  c7b07a9

  Copy full SHA for c7b07a9

• [NVIDIA] Fix an issue to use current stream for the nvfp4 quant (vllm-project#13632)

  kaixih

  authored
  27a09dc

  Copy full SHA for 27a09dc

• [Misc] Adding script to setup ray for multi-node vllm deployments (vllm-project#12913)

  Edwinhr716

  authored
  981f3c8

  Copy full SHA for 981f3c8

• Add llmaz as another integration (vllm-project#13643)

  Show description for 44c33f0
  kerthcet

  authored
  44c33f0

  Copy full SHA for 44c33f0

• [Neuron][Kernel] Vectorize KV cache load in FlashPagedAttention to maximize DMA bandwidth (vllm-project#13245)

  Show description for 3317008
  lingfanyu

  authored
  3317008

  Copy full SHA for 3317008

• [Bugfix] Fix max_num_batched_tokens for MLA (vllm-project#13620)

  Show description for 71face8
  mgoin

  authored
  71face8

  Copy full SHA for 71face8

Commits on Feb 20, 2025

• [Frontend] Add backend-specific options for guided decoding (vllm-project#13505)

  Show description for bfbc0b3
  joerunde

  authored
  bfbc0b3

  Copy full SHA for bfbc0b3

• fix neuron performance issue (vllm-project#13589)

  ajayvohra2005

  authored
  6a417b8

  Copy full SHA for 6a417b8

• [V1][Minor] Print KV cache size in token counts (vllm-project#13596)

  Show description for d3ea501
  WoosukKwon

  authored
  d3ea501

  Copy full SHA for d3ea501

• Update pre-commit's isort version to remove warnings (vllm-project#13614)

  hmellor

  authored
  34aad51

  Copy full SHA for 34aad51

Pagination

Previous

Next