Commits

Branch selector

 main

User selector

All users

Datepicker

All time

Commit History

Commits on Jan 28, 2025

• Replace missed warning_once for rerank API (#12472)

  Show description for 0f657bd
  mgoin

  authored
  0f657bd

  Copy full SHA for 0f657bd

• [V1][Metrics] Hook up IterationStats for Prometheus metrics (#12478)

  Show description for 3fd1fb6
  markmc

  authored
  3fd1fb6

  Copy full SHA for 3fd1fb6

• [Doc] Fix typo for x86 CPU installation (#12514)

  Show description for 925d2f1
  waltforme

  authored
  925d2f1

  Copy full SHA for 925d2f1

• [VLM] Merged multi-modal processor and V1 support for Qwen-VL (#12504)

  Show description for 8f58a51
  DarkLight1337

  authored
  8f58a51

  Copy full SHA for 8f58a51

• [Core] Make raw_request optional in ServingCompletion (#12503)

  Show description for 2079e43
  schoennenbeck

  authored
  2079e43

  Copy full SHA for 2079e43

• [V1] Include Engine Version in Logs (#12496)

  Show description for e29d435
  robertgshaw2-redhat

  authored
  e29d435

  Copy full SHA for e29d435

• Update README.md with V1 alpha release (#12495)

  ywang96

  authored
  8cbc424

  Copy full SHA for 8cbc424

• [CI] fix pre-commit error (#12494)

  Show description for dd66fd2
  MengqingCao

  authored
  dd66fd2

  Copy full SHA for dd66fd2

• [FEATURE] Enables offline /score for embedding models (#12021)

  Show description for 0f465ab
  gmarinho2

  authored
  0f465ab

  Copy full SHA for 0f465ab

• [CI/Build] Fixed the xla nightly issue report in #12451 (#12453)

  hosseinsarshar

  authored
  23a7cbc

  Copy full SHA for 23a7cbc

• Fix bad path in prometheus example (#12481)

  Show description for 426a5c3
  mgoin

  authored
  426a5c3

  Copy full SHA for 426a5c3

• [Neuron][Kernel] NKI-based flash-attention kernel with paged KV cache (#11277)

  Show description for ddee88d
  

  liangfu

  and

  JF-D

  authored
  ddee88d

  Copy full SHA for ddee88d

• Update pre-commit hooks (#12475)

  Show description for 823ab79
  hmellor

  authored
  823ab79

  Copy full SHA for 823ab79

Commits on Jan 27, 2025

• [Feature] [Spec decode]: Enable MLPSpeculator/Medusa and prompt_logprobs with ChunkedPrefill (#10132)

  Show description for 6116ca8
  

  NickLucche

  and

  wallashss

  authored
  6116ca8

  Copy full SHA for 6116ca8

• [FlashInfer] Upgrade to 0.2.0 (#11194)

  Show description for 2bc3fbb
  

  abmfy

  and

  youkaichao

  authored
  2bc3fbb

  Copy full SHA for 2bc3fbb

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

  Show description for 3f1fc74
  WoosukKwon

  authored
  3f1fc74

  Copy full SHA for 3f1fc74

• [V1][Metrics] Add initial Prometheus logger (#12416)

  Show description for 01ba927
  markmc

  authored
  01ba927

  Copy full SHA for 01ba927

• [Build] Only build 9.0a for scaled_mm and sparse kernels (#12339)

  Show description for 103bd17
  LucasWilkinson

  authored
  103bd17

  Copy full SHA for 103bd17

• [Bugfix] Fix gpt2 GGUF inference (#12467)

  Show description for ce69f7f
  Isotr0py

  authored
  ce69f7f

  Copy full SHA for ce69f7f

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

  Show description for 624a1e4
  WoosukKwon

  authored
  624a1e4

  Copy full SHA for 624a1e4

• [Bugfix] Fix missing seq_start_loc in xformers prefill metadata (#12464)

  Show description for 372bf08
  Isotr0py

  authored
  372bf08

  Copy full SHA for 372bf08

• [Bugfix] Fix Granite 3.0 MoE model loading (#12446)

  Show description for 5204ff5
  DarkLight1337

  authored
  5204ff5

  Copy full SHA for 5204ff5

• [Frontend] Support scores endpoint in run_batch (#12430)

  Show description for 0cc6b38
  pooyadavoodi

  authored
  0cc6b38

  Copy full SHA for 0cc6b38

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

  Show description for 28e0750
  WoosukKwon

  authored
  28e0750

  Copy full SHA for 28e0750

• [DOC] Add link to vLLM blog (#12460)

  Show description for 582cf78
  terrytangyuan

  authored
  582cf78

  Copy full SHA for 582cf78

• [Frontend] Rerank API (Jina- and Cohere-compatible API) (#12376)

  Show description for 0034b09
  K-Mistele

  authored
  0034b09

  Copy full SHA for 0034b09

Commits on Jan 26, 2025

• [Build/CI] Fix libcuda.so linkage (#12424)

  Show description for 72bac73
  tlrmchlsmth

  authored
  72bac73

  Copy full SHA for 72bac73

• [Bugfix][Kernel] Fix perf regression caused by PR #12405 (#12434)

  Show description for 68f1114
  LucasWilkinson

  authored
  68f1114

  Copy full SHA for 68f1114

• [Bugfix/CI] Fix broken kernels/test_mha.py (#12450)

  tlrmchlsmth

  authored
  72f4880

  Copy full SHA for 72f4880

• [Bugfix] Disable w16a16 2of4 sparse CompressedTensors24 (#12417)

  Show description for aa2cd2c
  

  tlrmchlsmth

  and

  mgoin

  authored
  aa2cd2c

  Copy full SHA for aa2cd2c

• [Frontend] generation_config.json for maximum tokens(#12242)

  Show description for 9ddc352
  

  9 peopleauthored
  9ddc352

  Copy full SHA for 9ddc352

• [Misc] Revert FA on ViT #12355 and #12435 (#12445)

  ywang96

  authored
  a525527

  Copy full SHA for a525527

• [V1][Bugfix] Fix assertion when mm hashing is turned off (#12439)

  Show description for 0ee349b
  ywang96

  authored
  0ee349b

  Copy full SHA for 0ee349b

• [V1][Perf] Reduce scheduling overhead in model runner after cuda sync (#12094)

  Show description for fa63e71
  youngkent

  authored
  fa63e71

  Copy full SHA for fa63e71

• [Misc][Bugfix] FA3 support to ViT MHA layer (#12435)

  Show description for 2a0309a
  

  ywang96

  and

  Isotr0py

  authored
  2a0309a

  Copy full SHA for 2a0309a

Pagination

Previous

Next