Commits

Branch selector

 b9c0605

User selector

All users

Datepicker

All time

Commit History

Commits on Jun 1, 2024

• [Feature][Kernel] Support bitsandbytes quantization and QLoRA (vllm-project#4776)

  chenqianfzh

  authored
  b9c0605

  Copy full SHA for b9c0605

• [Bugfix] Fix call to init_logger in openai server (vllm-project#4765)

  NadavShmayo

  authored
  37464a0

  Copy full SHA for 37464a0

• [Minor] Fix the path typo in loader.py: save_sharded_states.py -> save_sharded_state.py (vllm-project#5151)

  Show description for c354072
  dashanji

  authored
  c354072

  Copy full SHA for c354072

• [Kernel] Update Cutlass fp8 configs (vllm-project#5144)

  Show description for f081c3c
  

  3 peopleauthored
  f081c3c

  Copy full SHA for f081c3c

• [Kernel] Refactor CUTLASS kernels to always take scales that reside on the GPU (vllm-project#5137)

  tlrmchlsmth

  authored
  260d119

  Copy full SHA for 260d119

• [CI/Build] CMakeLists: build all extensions' cmake targets at the same time (vllm-project#5034)

  dtrifiro

  authored
  a360ff8

  Copy full SHA for a360ff8

• [Build] Guard against older CUDA versions when building CUTLASS 3.x kernels (vllm-project#5168)

  tlrmchlsmth

  authored
  1197e02

  Copy full SHA for 1197e02

• [Doc] Add checkmark for GPTBigCodeForCausalLM LoRA support (vllm-project#5171)

  njhill

  authored
  6575791

  Copy full SHA for 6575791

Commits on May 31, 2024

• [Model] Enable FP8 QKV in MoE and refine kernel tuning script (vllm-project#5039)

  comaniac

  authored
  e9899fb

  Copy full SHA for e9899fb

• [Misc]: optimize eager mode host time (vllm-project#4196)

  Show description for a377f0b
  

  FuncSherl

  and

  xuhao

  authored
  a377f0b

  Copy full SHA for a377f0b

• Revert "[Kernel] Marlin_24: Ensure the mma.sp instruction is using the ::ordered_metadata modifier (introduced with PTX 8.5)" (vllm-project#5149)

  simon-mo

  authored
  e9d3aa0

  Copy full SHA for e9d3aa0

• [Model] Support MAP-NEO model (vllm-project#5081)

  Show description for a22dea5
  

  xingweiqu

  and

  zhuohan123

  authored
  a22dea5

  Copy full SHA for a22dea5

• Fix cutlass sm_90a vesrion in CMakeList

  simon-mo

  committed
  533c217

  Copy full SHA for 533c217

• [Kernel] Marlin_24: Ensure the mma.sp instruction is using the ::ordered_metadata modifier (introduced with PTX 8.5) (vllm-project#5136)

  alexm-redhat

  authored
  6d21fa1

  Copy full SHA for 6d21fa1

• [Bugfix] Avoid Warnings in SparseML Activation Quantization (vllm-project#5120)

  robertgshaw2-redhat

  authored
  b35be54

  Copy full SHA for b35be54

Commits on May 30, 2024

• [Build] Disable sm_90a in cu11 (vllm-project#5141)

  simon-mo

  authored
  45a1a69

  Copy full SHA for 45a1a69

• Bump version to v0.4.3 (vllm-project#5046)

  simon-mo

  authored
  87a658c

  Copy full SHA for 87a658c

• add doc about serving option on dstack (vllm-project#3074)

  Show description for 429d897
  

  deep-diver

  and

  ywang96

  authored
  429d897

  Copy full SHA for 429d897

• [Doc] Use intersphinx and update entrypoints docs (vllm-project#5125)

  DarkLight1337

  authored
  a9bcc7a

  Copy full SHA for a9bcc7a

• [Misc] remove duplicate definition of seq_lens_tensor in model_runner.py (vllm-project#5129)

  ita9naiwa

  authored
  d79d9ea

  Copy full SHA for d79d9ea

• [CI/Build] increase wheel size limit to 200 MB (vllm-project#5130)

  youkaichao

  authored
  f758505

  Copy full SHA for f758505

• [Bugfix] Automatically Detect SparseML models (vllm-project#5119)

  robertgshaw2-redhat

  authored
  d910816

  Copy full SHA for d910816

• [BUGFIX] [FRONTEND] Correct chat logprobs (vllm-project#5029)

  Show description for 87d41c8
  br3no

  authored
  87d41c8

  Copy full SHA for 87d41c8

• [CI/Build] Docker cleanup functionality for amd servers (vllm-project#5112)

  Show description for e07aff9
  

  3 peopleauthored
  e07aff9

  Copy full SHA for e07aff9

• [Bugfix] gptq_marlin: Ensure g_idx_sort_indices is not a Parameter (vllm-project#5108)

  alexm-redhat

  authored
  5bf185a

  Copy full SHA for 5bf185a

Commits on May 29, 2024

• [Doc][Build] update after removing vllm-nccl (vllm-project#5103)

  Show description for 4fbcb0f
  

  youkaichao

  and

  ywang96

  authored
  4fbcb0f

  Copy full SHA for 4fbcb0f

• [Bugfix] logprobs is not compatible with the OpenAI spec vllm-project#4795 (vllm-project#5031)

  Etelis

  authored
  7c3604f

  Copy full SHA for 7c3604f

• [Core] Avoid the need to pass None values to Sequence.inputs (vllm-project#5099)

  DarkLight1337

  authored
  b1c2556

  Copy full SHA for b1c2556

• [Bugfix][CI/Build] Fix codespell failing to skip files in git diff (vllm-project#5097)

  DarkLight1337

  authored
  eb6c50c

  Copy full SHA for eb6c50c

• [Bugfix][CI/Build] Fix test and improve code for merge_async_iterators (vllm-project#5096)

  DarkLight1337

  authored
  eecd864

  Copy full SHA for eecd864

• [Doc]Replace deprecated flag in readme (vllm-project#4526)

  ronensc

  authored
  ae495c7

  Copy full SHA for ae495c7

• [Core] Cross-attention KV caching and memory-management (towards eventual encoder/decoder model support) (vllm-project#4837)

  afeldman-nm

  authored
  4238bc8

  Copy full SHA for 4238bc8

• [Core][Distributed] improve p2p access check (vllm-project#4992)

  youkaichao

  authored
  594392d

  Copy full SHA for 594392d

• [Bugfix] Fix arguments passed to Sequence in stop checker test (vllm-project#5092)

  DarkLight1337

  authored
  18c1f16

  Copy full SHA for 18c1f16

• [Core][Optimization] remove vllm-nccl (vllm-project#5091)

  youkaichao

  authored
  5bd3c65

  Copy full SHA for 5bd3c65

Pagination

Previous

Next