Commits

Branch selector

 35c4bc2

User selector

All users

Datepicker

All time

Commit History

Commits on Jan 12, 2024

• [Minor] Fix err msg (vllm-project#2431)

  WoosukKwon

  authored
  35c4bc2

  Copy full SHA for 35c4bc2

• Aligning top_p and top_k Sampling (vllm-project#1885)

  Show description for 218dc2c
  chenxu2048

  authored
  218dc2c

  Copy full SHA for 218dc2c

• Update quickstart.rst (vllm-project#2369)

  nautsimon

  authored
  827cbcd

  Copy full SHA for 827cbcd

• Suggest using dtype=half when OOM.

  chuanzhubin

  authored
  cb7a1c1

  Copy full SHA for cb7a1c1

• Address Phi modeling update 2 (vllm-project#2428)

  huiwy

  authored
  7878958

  Copy full SHA for 7878958

• Allow setting fastapi root_path argument (vllm-project#2341)

  chiragjn

  authored
  ce03624

  Copy full SHA for ce03624

• fix: deque mutated during iteration in abort_seq_group (vllm-project#2371)

  chenxu2048

  authored
  48cf1e4

  Copy full SHA for 48cf1e4

• Add gradio chatbot for openai webserver (vllm-project#2307)

  arkohut

  authored
  9746058

  Copy full SHA for 9746058

• [Minor] Fix the format in quick start guide related to Model Scope (vllm-project#2425)

  zhuohan123

  authored
  f745847

  Copy full SHA for f745847

• [DOC] Add additional comments for LLMEngine and AsyncLLMEngine (vllm-project#1011)

  litone01

  authored
  6549aef

  Copy full SHA for 6549aef

• Rename phi_1_5 -> phi (vllm-project#2385)

  WoosukKwon

  authored
  50376fa

  Copy full SHA for 50376fa

Commits on Jan 10, 2024

• get_ip(): Fix ipv4 ipv6 dualstack (vllm-project#2408)

  yunfeng-scale

  authored
  4b61c6b

  Copy full SHA for 4b61c6b

Commits on Jan 9, 2024

• [Speculative decoding 1/9] Optimized rejection sampler (vllm-project#2336)

  cadedaniel

  authored
  79d64c4

  Copy full SHA for 79d64c4

• Add baichuan chat template jinjia file (vllm-project#2390)

  taaaaaake

  authored
  74cd5ab

  Copy full SHA for 74cd5ab

Commits on Jan 8, 2024

• [Minor] Remove unused code in attention (vllm-project#2384)

  WoosukKwon

  authored
  28c3f12

  Copy full SHA for 28c3f12

• Fix eager mode performance (vllm-project#2377)

  WoosukKwon

  authored
  c884819

  Copy full SHA for c884819

Commits on Jan 7, 2024

• Changed scheduler to use deques instead of lists (vllm-project#2290)

  Show description for 05921a9
  

  NadavShmayo

  and

  WoosukKwon

  authored
  05921a9

  Copy full SHA for 05921a9

Commits on Jan 5, 2024

• Ensure metrics are logged regardless of requests (vllm-project#2347)

  ichernev

  authored
  d0215a5

  Copy full SHA for d0215a5

Commits on Jan 4, 2024

• Build docker image with shared objects from "build" step (vllm-project#2237)

  payoto

  authored
  937e7b7

  Copy full SHA for 937e7b7

• Miner fix of type hint (vllm-project#2340)

  beginlner

  authored
  aee8ef6

  Copy full SHA for aee8ef6

• Bump up to v0.2.7 (vllm-project#2337)

  WoosukKwon

  authored
  2e0b6e7

  Copy full SHA for 2e0b6e7

• Revert the changes in test_cache (vllm-project#2335)

  WoosukKwon

  authored
  9417671

  Copy full SHA for 9417671

Commits on Jan 3, 2024

• Remove unused const TIMEOUT_TO_PREVENT_DEADLOCK (vllm-project#2321)

  ronensc

  authored
  74d8d77

  Copy full SHA for 74d8d77

• Use NCCL instead of ray for control-plane communication to remove serialization overhead (vllm-project#2221)

  zhuohan123

  authored
  fd4ea8e

  Copy full SHA for fd4ea8e

• Remove deprecated parameter: concurrency_count (vllm-project#2315)

  ronensc

  authored
  1066cbd

  Copy full SHA for 1066cbd

• Enable CUDA graph for GPTQ & SqueezeLLM (vllm-project#2318)

  WoosukKwon

  authored
  6ef00b0

  Copy full SHA for 6ef00b0

• [Minor] Fix typo and remove unused code (vllm-project#2305)

  esmeetu

  authored
  9140561

  Copy full SHA for 9140561

• [FIX] Support non-zero CUDA devices in custom kernels (vllm-project#1959)

  jeejeelee

  authored
  77af974

  Copy full SHA for 77af974

Commits on Dec 30, 2023

• Support GPT-NeoX Models without attention biases (vllm-project#2301)

  dalgarak

  authored
  4934d49

  Copy full SHA for 4934d49

Commits on Dec 27, 2023

• [BUGFIX] Fix communication test (vllm-project#2285)

  zhuohan123

  authored
  358c328

  Copy full SHA for 358c328

Commits on Dec 26, 2023

• [BUGFIX] Fix the path of test prompts (vllm-project#2273)

  zhuohan123

  authored
  4aaafdd

  Copy full SHA for 4aaafdd

• [BUGFIX] Fix API server test (vllm-project#2270)

  zhuohan123

  authored
  66b108d

  Copy full SHA for 66b108d

• [BUGFIX] Do not return ignored sentences twice in async llm engine (vllm-project#2258)

  zhuohan123

  authored
  e0ff920

  Copy full SHA for e0ff920

• [Docs] Add "About" Heading to README.md (vllm-project#2260)

  blueceiling

  authored
  face83c

  Copy full SHA for face83c

Commits on Dec 23, 2023

• [Docs] Update installation instructions to include CUDA 11.8 xFormers (vllm-project#2246)

  skt7

  authored
  1db83e3

  Copy full SHA for 1db83e3

Pagination

Previous

Next