Commits

Branch selector

 main

User selector

All users

Datepicker

All time

Commit History

Commits on Feb 3, 2025

• [MISC] Remove model input dumping when exception (vllm-project#12582)

  Show description for cf58b9c
  comaniac

  authored
  cf58b9c

  Copy full SHA for cf58b9c

• [Model] Add Deepseek V3 fp8_w8a8 configs for B200 (vllm-project#12707)

  kushanam

  authored
  4797dad

  Copy full SHA for 4797dad

• Squelch MLA warning for Compressed-Tensors Models (vllm-project#12704)

  Show description for 6dd5e52
  kylesayrs

  authored
  6dd5e52

  Copy full SHA for 6dd5e52

• [Bugfix][Kernel] Fix per-token/per-channel quantization for Hopper scaled mm (vllm-project#12696)

  Show description for c11de33
  tlrmchlsmth

  authored
  c11de33

  Copy full SHA for c11de33

• [Misc] Fix improper placement of SPDX header in scripts (vllm-project#12694)

  Show description for 33e0602
  russellb

  authored
  33e0602

  Copy full SHA for 33e0602

• [Model]: Add transformers backend support (vllm-project#11330)

  Show description for a1a2aaa
  

  5 peopleauthored
  a1a2aaa

  Copy full SHA for a1a2aaa

• [ci/build] fix gh200 test (vllm-project#12681)

  Show description for 1298a40
  youkaichao

  authored
  1298a40

  Copy full SHA for 1298a40

• [cuda] manually import the correct pynvml module (vllm-project#12679)

  Show description for ad4a9dc
  youkaichao

  authored
  ad4a9dc

  Copy full SHA for ad4a9dc

• Fix for attention layers to remain unquantized during moe_wn16 quant (vllm-project#12570)

  Show description for b998645
  

  31 peopleauthored
  b998645

  Copy full SHA for b998645

• Properly check if all fused layers are in the list of targets (vllm-project#12666)

  Show description for c5932e5
  eldarkurtic

  authored
  c5932e5

  Copy full SHA for c5932e5

• make sure mistral_common not imported for non-mistral models (vllm-project#12669)

  Show description for 20579c0
  youkaichao

  authored
  20579c0

  Copy full SHA for 20579c0

• [Kernel] port sgl moe_align_block_size kernels (vllm-project#12574)

  Show description for 95460fc
  chenyang78

  authored
  95460fc

  Copy full SHA for 95460fc

• [Doc] Deprecate Discord (vllm-project#12668)

  zhuohan123

  authored
  326fcc8

  Copy full SHA for 326fcc8

• [doc][misc] clarify VLLM_HOST_IP for multi-node inference (vllm-project#12667)

  Show description for e643309
  youkaichao

  authored
  e643309

  Copy full SHA for e643309

Commits on Feb 2, 2025

• [Misc] Add SPDX-License-Identifier headers to python source files (vllm-project#12628)

  Show description for e489ad7
  russellb

  authored
  e489ad7

  Copy full SHA for e489ad7

• [Hardware][Intel GPU] add XPU bf16 support (vllm-project#12392)

  Show description for f256ebe
  jikunshang

  authored
  f256ebe

  Copy full SHA for f256ebe

• [Core][v1] Unify allocating slots in prefill and decode in KV cache manager (vllm-project#12608)

  Show description for f8ece6e
  ShawnD200

  authored
  f8ece6e

  Copy full SHA for f8ece6e

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

  Show description for abfcdcd
  WoosukKwon

  authored
  abfcdcd

  Copy full SHA for abfcdcd

• [Core] Silence unnecessary deprecation warnings (vllm-project#12620)

  Show description for e497f33
  russellb

  authored
  e497f33

  Copy full SHA for e497f33

• [Bugfix] fix moe_wna16 get_quant_method (vllm-project#12648)

  Show description for baaa2b2
  jinzhen-lin

  authored
  baaa2b2

  Copy full SHA for baaa2b2

Commits on Feb 1, 2025

• doc: fixing minor typo in readme.md (vllm-project#12643)

  Show description for b4e5c03
  vicenteherrera

  authored
  b4e5c03

  Copy full SHA for b4e5c03

• Apply torch.compile to fused_moe/grouped_topk (vllm-project#12637)

  mgoin

  authored
  3194039

  Copy full SHA for 3194039

• Disable chunked prefill and/or prefix caching when MLA is enabled (vllm-project#12642)

  Show description for 4f4d427
  

  simon-mo

  and

  mgoin

  authored
  4f4d427

  Copy full SHA for 4f4d427

• [CI/Build] Add label automation for structured-output, speculative-decoding, v1 (vllm-project#12280)

  Show description for 1e36983
  russellb

  authored
  1e36983

  Copy full SHA for 1e36983

• [Attention] Deepseek v3 MLA support with FP8 compute (vllm-project#12601)

  Show description for baeded2
  

  7 peopleauthored
  baeded2

  Copy full SHA for baeded2

• Fix: Respect sparsity_config.ignore in Cutlass Integration (vllm-project#12517)

  Show description for 3e1c76c
  rahul-tuli

  authored
  3e1c76c

  Copy full SHA for 3e1c76c

• [Bugfix/CI] Fixup benchmark_moe.py (vllm-project#12562)

  Show description for cfa134d
  tlrmchlsmth

  authored
  cfa134d

  Copy full SHA for cfa134d

• [ci] Upgrade transformers to 4.48.2 in CI dependencies (vllm-project#12599)

  khluu

  authored
  35b7a05

  Copy full SHA for 35b7a05

• Fix target matching for fused layers with compressed-tensors (vllm-project#12617)

  Show description for 1867c25
  eldarkurtic

  authored
  1867c25

  Copy full SHA for 1867c25

• [BugFix] fix wrong output when using lora and num_scheduler_steps=8 (vllm-project#11161)

  Show description for cb3e73e
  

  3 peopleauthored
  cb3e73e

  Copy full SHA for cb3e73e

• [V1] Bugfix: Validate Model Input Length (vllm-project#12600)

  Show description for b1340f9
  robertgshaw2-redhat

  authored
  b1340f9

  Copy full SHA for b1340f9

Commits on Jan 31, 2025

• [Doc] int4 w4a16 example (vllm-project#12585)

  Show description for 44bbca7
  

  brian-dellabetta

  and

  mgoin

  authored
  44bbca7

  Copy full SHA for 44bbca7

• [Doc] Improve installation signposting (vllm-project#12575)

  Show description for 60808bd
  hmellor

  authored
  60808bd

  Copy full SHA for 60808bd

• [Feature] Fix guided decoding blocking bitmask memcpy (vllm-project#12563)

  Show description for fc54214
  xpbowler

  authored
  fc54214

  Copy full SHA for fc54214

• [Kernel][Quantization] Integrate block-quantized CUTLASS kernels for DeepSeekV3 (vllm-project#12587)

  Show description for eb5741a
  tlrmchlsmth

  authored
  eb5741a

  Copy full SHA for eb5741a

Pagination

Previous

Next