Commits

Branch selector

 main

User selector

All users

All time

Commit History

Commits on Feb 21, 2024

• fix get_ip/get_open_port failure in pure ipv6 environment

  Jingru

  committed
  2675456

  Copy full SHA for 2675456

• fix get_ip/get_open_port failure in pure ipv6 environment

  Jingru

  committed
  bf4adc4

  Copy full SHA for bf4adc4

Commits on Feb 20, 2024

• [ROCm] include gfx908 as supported (#2792)

  jamestwhedbee

  authored
  264017a

  Copy full SHA for 264017a

Commits on Feb 19, 2024

• Fix `vllm:prompt_tokens_total` metric calculation (#2869)

  ronensc

  authored
  e433c11

  Copy full SHA for e433c11

• Add warning to prevent changes to benchmark api server (#2858)

  simon-mo

  authored
  86fd8bb

  Copy full SHA for 86fd8bb

• Support OLMo models. (#2832)

  Isotr0py

  authored
  ab3a5a8

  Copy full SHA for ab3a5a8

• [Test] Add basic correctness test (#2908)

  zhuohan123

  authored
  a61f052

  Copy full SHA for a61f052

Commits on Feb 18, 2024

• [Minor] Small fix to make distributed init logic in worker looks cleaner (#2905)

  zhuohan123

  authored
  537c975

  Copy full SHA for 537c975

• Add code-revision config argument for Hugging Face Hub (#2892)

  mbm-ai

  authored
  786b7f1

  Copy full SHA for 786b7f1

Commits on Feb 17, 2024

• multi-LoRA as extra models in OpenAI server (#2775)

  Show description for 8f36444
  jvmncs

  authored
  8f36444

  Copy full SHA for 8f36444

• Defensively copy `sampling_params` (#2881)

  Show description for 185b2c2
  njhill

  authored
  185b2c2

  Copy full SHA for 185b2c2

Commits on Feb 16, 2024

• Bump up to v0.3.1 (#2887)

  WoosukKwon

  authored
  5f08050

  Copy full SHA for 5f08050

• Prefix Caching- fix t4 triton error (#2517)

  caoshiyi

  authored
  64da65b

  Copy full SHA for 64da65b

Commits on Feb 15, 2024

• [ROCm] Dockerfile fix for flash-attention build (#2885)

  hongxiayang

  authored
  5255d99

  Copy full SHA for 5255d99

• Fix DeciLM (#2883)

  pcmoritz

  authored
  4f2ad11

  Copy full SHA for 4f2ad11

• [BugFix] Fix GC bug for `LLM` class (#2882)

  WoosukKwon

  authored
  d7afab6

  Copy full SHA for d7afab6

• Align LoRA code between Mistral and Mixtral (fixes #2875) (#2880)

  Show description for 31348df
  

  pcmoritz

  and

  Pernekhan

  authored
  31348df

  Copy full SHA for 31348df

Commits on Feb 14, 2024

• Don't use cupy NCCL for AMD backends (#2855)

  WoosukKwon

  authored
  25e86b6

  Copy full SHA for 25e86b6

• Migrate AquilaForCausalLM to LlamaForCausalLM (#2867)

  esmeetu

  authored
  4efbac6

  Copy full SHA for 4efbac6

• Fix docker python version (#2845)

  NikolaBorisov

  authored
  87069cc

  Copy full SHA for 87069cc

• [Fix] Fix memory profiling when GPU is used by multiple processes (#2863)

  WoosukKwon

  authored
  7e45107

  Copy full SHA for 7e45107

• Fix internlm after https://github.com/vllm-project/vllm/pull/2860 (#2861)

  pcmoritz

  authored
  0c48b37

  Copy full SHA for 0c48b37

• Migrate InternLMForCausalLM to LlamaForCausalLM (#2860)

  Show description for 7eacffd
  

  pcmoritz

  and

  esmeetu

  authored
  7eacffd

  Copy full SHA for 7eacffd

Commits on Feb 13, 2024

• Add LoRA support for Mixtral (#2831)

  Show description for 2a543d6
  tterrysun

  authored
  2a543d6

  Copy full SHA for 2a543d6

• Remove Yi model definition, please use `LlamaForCausalLM` instead (#2854)

  Show description for 317b29d
  

  pcmoritz

  and

  esmeetu

  authored
  317b29d

  Copy full SHA for 317b29d

• Use CuPy for CUDA graphs (#2811)

  WoosukKwon

  authored
  a463c33

  Copy full SHA for a463c33

• Revert "Refactor llama family models (#2637)" (#2851)

  Show description for ea35600
  pcmoritz

  authored
  ea35600

  Copy full SHA for ea35600

• Refactor llama family models (#2637)

  esmeetu

  authored
  5c976a7

  Copy full SHA for 5c976a7

• [CI] Ensure documentation build is checked in CI (#2842)

  simon-mo

  authored
  f964493

  Copy full SHA for f964493

• Serving Benchmark Refactoring (#2433)

  ywang96

  authored
  a4211a4

  Copy full SHA for a4211a4

Commits on Feb 12, 2024

• Refactor 2 awq gemm kernels into m16nXk32 (#2723)

  Show description for 5638364
  

  zcnrex

  and

  Chunan Zeng

  authored
  5638364

  Copy full SHA for 5638364

• Add documentation section about LoRA (#2834)

  pcmoritz

  authored
  4ca2c35

  Copy full SHA for 4ca2c35

Commits on Feb 11, 2024

• [ROCm] support Radeon™ 7900 series (gfx1100) without using flash-attention (#2768)

  hongxiayang

  authored
  0580aab

  Copy full SHA for 0580aab

Commits on Feb 8, 2024

• Disable custom all reduce by default (#2808)

  WoosukKwon

  authored
  3711811

  Copy full SHA for 3711811

• [Ray] Integration compiled DAG off by default (#2471)

  rkooo567

  authored
  65b89d1

  Copy full SHA for 65b89d1