Commits

Branch selector

 main

User selector

WoosukKwon

Datepicker

All time

Commit History

Commits on Feb 11, 2025

• [Benchmark] Add BurstGPT to benchmark_serving (vllm-project#13063)

  Show description for 58047c6
  

  WoosukKwon

  and

  ywang96

  authored
  58047c6

  Copy full SHA for 58047c6

• [V1][Minor] Move scheduler outputs to a separate file (vllm-project#13062)

  Show description for 2ff4857
  WoosukKwon

  authored
  2ff4857

  Copy full SHA for 2ff4857

Commits on Feb 8, 2025

• [V1] Cache uses_mrope in GPUModelRunner (vllm-project#12969)

  WoosukKwon

  authored
  24700c3

  Copy full SHA for 24700c3

• [V1][Minor] Remove outdated comment (vllm-project#12968)

  Show description for 870c374
  WoosukKwon

  authored
  870c374

  Copy full SHA for 870c374

• [V1][Minor] Move cascade attn logic outside _prepare_inputs (vllm-project#12943)

  Show description for 4ea48fb
  WoosukKwon

  authored
  4ea48fb

  Copy full SHA for 4ea48fb

• [V1] Move KV block hashes from Request to KVCacheManager (vllm-project#12922)

  Show description for 3243158
  WoosukKwon

  authored
  3243158

  Copy full SHA for 3243158

• [V1][Minor] Remove outdated comment (vllm-project#12928)

  Show description for b21f0f9
  WoosukKwon

  authored
  b21f0f9

  Copy full SHA for b21f0f9

Commits on Feb 4, 2025

• [V1] Remove scheduling constraint on partial requests (vllm-project#12674)

  Show description for 18a88fc
  WoosukKwon

  authored
  18a88fc

  Copy full SHA for 18a88fc

Commits on Feb 2, 2025

• [V1][Minor] Avoid frequently creating ConstantList (vllm-project#12653)

  Show description for abfcdcd
  WoosukKwon

  authored
  abfcdcd

  Copy full SHA for abfcdcd

Commits on Jan 29, 2025

• [V1][BugFix] Free encoder cache for aborted requests (vllm-project#12545)

  Show description for e0cc5f2
  WoosukKwon

  authored
  e0cc5f2

  Copy full SHA for e0cc5f2

Commits on Jan 27, 2025

• [V1][CI/Test] Do basic test for top-p & top-k sampling (vllm-project#12469)

  Show description for 3f1fc74
  WoosukKwon

  authored
  3f1fc74

  Copy full SHA for 3f1fc74

• [V1][Minor] Minor optimizations for update_from_output (vllm-project#12454)

  Show description for 624a1e4
  WoosukKwon

  authored
  624a1e4

  Copy full SHA for 624a1e4

• [V1] Avoid list creation in input preparation (vllm-project#12457)

  Show description for 28e0750
  WoosukKwon

  authored
  28e0750

  Copy full SHA for 28e0750

Commits on Jan 24, 2025

• [V1] Increase default batch size for H100/H200 (vllm-project#12369)

  Show description for 0e74d79
  WoosukKwon

  authored
  0e74d79

  Copy full SHA for 0e74d79

Commits on Jan 23, 2025

• [Docs] Add meetup slides (vllm-project#12345)

  Show description for 682b55b
  WoosukKwon

  authored
  682b55b

  Copy full SHA for 682b55b

Commits on Jan 15, 2025

• [V1][BugFix] Fix edge case in VLM scheduling (vllm-project#12065)

  Show description for b7ee940
  WoosukKwon

  authored
  b7ee940

  Copy full SHA for b7ee940

Commits on Jan 14, 2025

• [Docs] Add Sky Computing Lab to project intro (vllm-project#12019)

  Show description for 1a40125
  WoosukKwon

  authored
  1a40125

  Copy full SHA for 1a40125

Commits on Jan 6, 2025

• [V1] Add BlockTable class (vllm-project#11693)

  Show description for 06bfb51
  WoosukKwon

  authored
  06bfb51

  Copy full SHA for 06bfb51

Commits on Jan 2, 2025

• [V1][Minor] Optimize token_ids_cpu copy (vllm-project#11692)

  Show description for b55ed6e
  WoosukKwon

  authored
  b55ed6e

  Copy full SHA for b55ed6e

Commits on Jan 1, 2025

• [V1] Implement Cascade Attention (vllm-project#11635)

  Show description for 7300144
  WoosukKwon

  authored
  7300144

  Copy full SHA for 7300144

Commits on Dec 28, 2024

• [V1][Minor] Set pin_memory=False for token_ids_cpu tensor (vllm-project#11581)

  Show description for 42bb201
  WoosukKwon

  authored
  42bb201

  Copy full SHA for 42bb201

Commits on Dec 27, 2024

• [V1] Fix yapf (vllm-project#11538)

  Show description for 81b979f
  WoosukKwon

  authored
  81b979f

  Copy full SHA for 81b979f

• [V1] Use FlashInfer Sampling Kernel for Top-P & Top-K Sampling (vllm-project#11394)

  Show description for 371d04d
  WoosukKwon

  authored
  371d04d

  Copy full SHA for 371d04d

Commits on Dec 21, 2024

• [V1][Bugfix] Skip hashing empty or None mm_data (vllm-project#11386)

  Show description for dd2b563
  WoosukKwon

  authored
  dd2b563

  Copy full SHA for dd2b563

Commits on Dec 15, 2024

• [V1][Minor] Cache np arange to reduce input preparation overhead (vllm-project#11214)

  Show description for 25ebed2
  WoosukKwon

  authored
  25ebed2

  Copy full SHA for 25ebed2

Commits on Dec 12, 2024

• [V1] Use more persistent buffers to optimize input preparation overheads (vllm-project#11111)

  Show description for f092153
  WoosukKwon

  authored
  f092153

  Copy full SHA for f092153

Commits on Dec 11, 2024

• [V1] Use input_ids as input for text-only models (vllm-project#11032)

  Show description for d643c2a
  WoosukKwon

  authored
  d643c2a

  Copy full SHA for d643c2a

Commits on Dec 10, 2024

• [V1][Bugfix] Always set enable_chunked_prefill = True for V1 (vllm-project#11061)

  Show description for 134810b
  WoosukKwon

  authored
  134810b

  Copy full SHA for 134810b

Commits on Dec 9, 2024

• [V1] Do not store None in self.generators (vllm-project#11038)

  Show description for 6faec54
  WoosukKwon

  authored
  6faec54

  Copy full SHA for 6faec54

• [V1] Further reduce CPU overheads in flash-attn (vllm-project#10989)

  Show description for 3b61cb4
  WoosukKwon

  authored
  3b61cb4

  Copy full SHA for 3b61cb4

Commits on Dec 5, 2024

• [V1] Fix when max_model_len is not divisible by block_size (vllm-project#10903)

  Show description for 2a56e12
  WoosukKwon

  authored
  2a56e12

  Copy full SHA for 2a56e12

Commits on Dec 2, 2024

• [Kernel] Use out arg in flash_attn_varlen_func (vllm-project#10811)

  Show description for 073a4bd
  WoosukKwon

  authored
  073a4bd

  Copy full SHA for 073a4bd

Commits on Nov 28, 2024

• [V1] Optimize the CPU overheads in FlashAttention custom op (vllm-project#10733)

  Show description for 98f47f2
  WoosukKwon

  authored
  98f47f2

  Copy full SHA for 98f47f2

• [Kernel] Update vllm-flash-attn version to reduce CPU overheads (vllm-project#10742)

  Show description for 8c1e77f
  WoosukKwon

  authored
  8c1e77f

  Copy full SHA for 8c1e77f

• [Kernel] Update vllm-flash-attn version (vllm-project#10736)

  Show description for 9a8bff0
  WoosukKwon

  authored
  9a8bff0

  Copy full SHA for 9a8bff0

Pagination

Previous

Next