Commits

Branch selector

 main

User selector

esmeetu

Datepicker

All time

Commit History

Commits on May 1, 2024

• [Bugfix][Core] Fix and refactor logging stats (vllm-project#4336)

  esmeetu

  authored
  3a922c1

  Copy full SHA for 3a922c1

Commits on Apr 27, 2024

• [Bugfix][Core] Fix get decoding config from ray (vllm-project#4335)

  esmeetu

  authored
  7134303

  Copy full SHA for 7134303

• [Misc] Fix logger format typo (vllm-project#4396)

  esmeetu

  authored
  87f545b

  Copy full SHA for 87f545b

Commits on Apr 25, 2024

• [Misc] Fix flash attention backend log (vllm-project#4368)

  esmeetu

  authored
  b6dcb4d

  Copy full SHA for b6dcb4d

Commits on Apr 15, 2024

• [Misc] Upgrade triton to 2.2.0 (vllm-project#4061)

  esmeetu

  authored
  8db1bf3

  Copy full SHA for 8db1bf3

Commits on Apr 8, 2024

• [BugFix][Model] Fix commandr RoPE max_position_embeddings (vllm-project#3919)

  esmeetu

  authored
  d036198

  Copy full SHA for d036198

Commits on Mar 29, 2024

• [BugFix][Frontend] Fix completion logprobs=0 error (vllm-project#3731)

  esmeetu

  authored
  f510395

  Copy full SHA for f510395

• [BugFix] Fix tokenizer out of vocab size (vllm-project#3685)

  esmeetu

  authored
  6110c39

  Copy full SHA for 6110c39

Commits on Mar 28, 2024

• [Core] Support multi-node inference(eager and cuda graph) (vllm-project#3686)

  esmeetu

  authored
  515386e

  Copy full SHA for 515386e

• [Model] Fix and clean commandr (vllm-project#3671)

  esmeetu

  authored
  10e6322

  Copy full SHA for 10e6322

Commits on Mar 22, 2024

• [Bugfix][Model] Fix Qwen2 (vllm-project#3554)

  esmeetu

  authored
  ea5f14e

  Copy full SHA for ea5f14e

Commits on Mar 21, 2024

• [Misc][Log] Add log for tokenizer length not equal to vocabulary size (vllm-project#3500)

  esmeetu

  authored
  8657323

  Copy full SHA for 8657323

Commits on Mar 20, 2024

• Migrate logits computation and gather to model_runner (vllm-project#3233)

  esmeetu

  authored
  f1c0fc3

  Copy full SHA for f1c0fc3

Commits on Mar 11, 2024

• [BugFix] Fix get tokenizer when using ray (vllm-project#3301)

  esmeetu

  authored
  9e8744a

  Copy full SHA for 9e8744a

Commits on Feb 27, 2024

• Fix stablelm (vllm-project#3038)

  esmeetu

  authored
  4dd6416

  Copy full SHA for 4dd6416

• Don't use cupy when enforce_eager=True (vllm-project#3037)

  esmeetu

  authored
  c1c0d00

  Copy full SHA for c1c0d00

• [Minor] Remove unused config files (vllm-project#3039)

  esmeetu

  authored
  d9f726c

  Copy full SHA for d9f726c

Commits on Feb 22, 2024

• Migrate MistralForCausalLM to LlamaForCausalLM (vllm-project#2868)

  esmeetu

  authored
  344020c

  Copy full SHA for 344020c

Commits on Feb 14, 2024

• Migrate AquilaForCausalLM to LlamaForCausalLM (vllm-project#2867)

  esmeetu

  authored
  4efbac6

  Copy full SHA for 4efbac6

Commits on Feb 13, 2024

• Refactor llama family models (vllm-project#2637)

  esmeetu

  authored
  5c976a7

  Copy full SHA for 5c976a7

Commits on Jan 20, 2024

• [Bugfix] fix load local safetensors model (vllm-project#2512)

  esmeetu

  authored
  91a61da

  Copy full SHA for 91a61da

Commits on Jan 14, 2024

• [Minor] Optimize cuda graph memory usage (vllm-project#2437)

  esmeetu

  authored
  9f659bf

  Copy full SHA for 9f659bf

Commits on Jan 3, 2024

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

  esmeetu

  authored
  9140561

  Copy full SHA for 9140561

Commits on Dec 16, 2023

• Simplify weight loading logic (vllm-project#2133)

  esmeetu

  authored
  eed74a5

  Copy full SHA for eed74a5

Commits on Dec 3, 2023

• add custom server params (vllm-project#1868)

  esmeetu

  authored
  60dc62d

  Copy full SHA for 60dc62d

Commits on Nov 30, 2023

• Fix rope cache key error (vllm-project#1867)

  esmeetu

  authored
  d27f4ba

  Copy full SHA for d27f4ba

Commits on Nov 18, 2023

• Support Min P Sampler (vllm-project#1642)

  esmeetu

  authored
  e87557b

  Copy full SHA for e87557b

Commits on Nov 6, 2023

• Support Yi model (vllm-project#1567)

  esmeetu

  authored
  e7f579e

  Copy full SHA for e7f579e

Commits on Sep 21, 2023

• clean api code, remove redundant background task. (vllm-project#1102)

  esmeetu

  authored
  2d1e86f

  Copy full SHA for 2d1e86f

Commits on Sep 18, 2023

• align llm_engine and async_engine. (vllm-project#1081)

  esmeetu

  authored
  95592fa

  Copy full SHA for 95592fa

Commits on Jul 14, 2023

• add vocab padding for LLama(Support WizardLM) (vllm-project#411)

  esmeetu

  authored
  7b6ae94

  Copy full SHA for 7b6ae94