Commits

Branch selector

 main

User selector

All users

Datepicker

All time

Commit History

Commits on Sep 18, 2023

• [FIX] Don't initialize parameter by default (vllm-project#1067)

  zhuohan123

  authored
  90979c3

  Copy full SHA for 90979c3

Commits on Sep 17, 2023

• Fix hanging when prompt exceeds limit (vllm-project#1029)

  chenxu2048

  authored
  e21d768

  Copy full SHA for e21d768

• Remove AsyncLLMEngine busy loop, shield background task (vllm-project#1059)

  Yard1

  authored
  ff36139

  Copy full SHA for ff36139

Commits on Sep 16, 2023

• Implement AWQ quantization support for LLaMA (vllm-project#1032)

  Show description for e3e79e9
  

  5 peopleauthored
  e3e79e9

  Copy full SHA for e3e79e9

Commits on Sep 15, 2023

• Abort when coroutine is cancelled (vllm-project#1020)

  rucyang

  authored
  b9fe461

  Copy full SHA for b9fe461

• Fix warning message on LLaMA FastTokenizer (vllm-project#1037)

  WoosukKwon

  authored
  64ca424

  Copy full SHA for 64ca424

• Only fail if logit_bias has actual values (vllm-project#1045)

  LLukas22

  authored
  b5f93d0

  Copy full SHA for b5f93d0

• Add pandas to requirements.txt (vllm-project#1047)

  Show description for a589369
  WoosukKwon

  authored
  a589369

  Copy full SHA for a589369

Commits on Sep 14, 2023

• Fix detokenization leaving special tokens (vllm-project#1044)

  Show description for dd54a4b
  Yard1

  authored
  dd54a4b

  Copy full SHA for dd54a4b

• Announce paper release (vllm-project#1036)

  WoosukKwon

  authored
  eda1a7c

  Copy full SHA for eda1a7c

Commits on Sep 13, 2023

• [FIX] Minor bug fixes (vllm-project#1035)

  Show description for f04908c
  zhuohan123

  authored
  f04908c

  Copy full SHA for f04908c

• Add Model Revision Support (vllm-project#1014)

  Show description for ab019ee
  

  3 peopleauthored
  ab019ee

  Copy full SHA for ab019ee

• Use TGI-like incremental detokenization (vllm-project#984)

  Yard1

  authored
  9841d48

  Copy full SHA for 9841d48

• Fix typo in README.md (vllm-project#1033)

  eltociear

  authored
  3272d7a

  Copy full SHA for 3272d7a

Commits on Sep 12, 2023

• Make max_model_len configurable (vllm-project#972)

  Yard1

  authored
  0bb1e88

  Copy full SHA for 0bb1e88

• add option to shorten prompt print in log (vllm-project#991)

  Show description for d6545ad
  

  3 peopleauthored
  d6545ad

  Copy full SHA for d6545ad

Commits on Sep 11, 2023

• Bump up the version to v0.1.7 (vllm-project#1013)

  WoosukKwon

  authored
  90eb3f4

  Copy full SHA for 90eb3f4

• Use FP32 in RoPE initialization (vllm-project#1004)

  Show description for e67b4f2
  

  WoosukKwon

  and

  imoneoi

  authored
  e67b4f2

  Copy full SHA for e67b4f2

• Update setup.py (vllm-project#1006)

  WoosukKwon

  authored
  d6770d1

  Copy full SHA for d6770d1

Commits on Sep 10, 2023

• [Docs] Update installation page (vllm-project#1005)

  WoosukKwon

  authored
  b9cecc2

  Copy full SHA for b9cecc2

• fix: CUDA error when inferencing with Falcon-40B base model (vllm-project#992)

  kyujin-cho

  authored
  898285c

  Copy full SHA for 898285c

Commits on Sep 9, 2023

• Fix wrong dtype in PagedAttentionWithALiBi bias (vllm-project#996)

  Show description for a62de9e
  Yard1

  authored
  a62de9e

  Copy full SHA for a62de9e

• fix "tansformers_module" ModuleNotFoundError when load model with trust_remote_code=True (vllm-project#871)

  Jingru

  authored
  4042d19

  Copy full SHA for 4042d19

Commits on Sep 8, 2023

• Bump up the version to v0.1.6 (vllm-project#989)

  zhuohan123

  authored
  1117aa1

  Copy full SHA for 1117aa1

• Start background task in AsyncLLMEngine.generate (vllm-project#988)

  Show description for 0804384
  

  Yard1

  and

  zhuohan123

  authored
  0804384

  Copy full SHA for 0804384

• faster startup of vLLM (vllm-project#982)

  Show description for 4b5bcf8
  

  ri938

  and

  robirv938

  authored
  4b5bcf8

  Copy full SHA for 4b5bcf8

Commits on Sep 7, 2023

• Bump up the version to v0.1.5 (vllm-project#944)

  WoosukKwon

  authored
  852ef5b

  Copy full SHA for 852ef5b

• [FIX] Fix Alibi implementation in PagedAttention kernel (vllm-project#945)

  Show description for db09d4a
  

  3 peopleauthored
  db09d4a

  Copy full SHA for db09d4a

• Enable safetensors loading for all models (vllm-project#974)

  zhuohan123

  authored
  c957c74

  Copy full SHA for c957c74

• Make AsyncLLMEngine more robust & fix batched abort (vllm-project#969)

  Show description for c07ece5
  

  Yard1

  and

  avnishn

  authored
  c07ece5

  Copy full SHA for c07ece5

• Bum up transformers version (vllm-project#976)

  WoosukKwon

  authored
  7a9c20c

  Copy full SHA for 7a9c20c

• Set torch default dtype in a context manager (vllm-project#971)

  Show description for 005ba45
  Yard1

  authored
  005ba45

  Copy full SHA for 005ba45

Commits on Sep 6, 2023

• [BugFix] Implement RoPE for GPT-J (vllm-project#941)

  WoosukKwon

  authored
  320a622

  Copy full SHA for 320a622

• Use queue for finished requests (vllm-project#957)

  Yard1

  authored
  c9927c1

  Copy full SHA for c9927c1

Commits on Sep 5, 2023

• Clean up kernel unit tests (vllm-project#938)

  WoosukKwon

  authored
  fbd80ad

  Copy full SHA for fbd80ad

Pagination

Previous

Next