Commits

Branch selector

 main

User selector

All users

Datepicker

All time

Commit History

Commits on May 20, 2025

• New cache tests and refactored Hybrid Cache (#37972)

  manueldeprada

  authored
  d34e21e

  Copy full SHA for d34e21e

• Add `Llama4TextModel` to `AutoModel` mapping (#38162)

  Show description for 183fb36
  

  Matthew Hoffman

  authored
  183fb36

  Copy full SHA for 183fb36

• Remove trust_remote_code=True tests from bnb quantization tests (MPT now integrated) (#38206)

  Show description for f022bf9
  Titus-von-Koeller

  authored
  f022bf9

  Copy full SHA for f022bf9

• [fix] sliding window attention mask (#38045)

  Show description for 0a52bd2
  

  zucchini-nlp

  and

  gante

  authored
  0a52bd2

  Copy full SHA for 0a52bd2

• Fix broken example generation script for Llama3 (#38062)

  Show description for 555715f
  sarckk

  authored
  555715f

  Copy full SHA for 555715f

• Fix: make docs work better with doc builder (#38213)

  S1ro1

  authored
  7a611f0

  Copy full SHA for 7a611f0

• enable misc cases on XPU & use device agnostic APIs for cases in tests (#38192)

  Show description for 3bd1c20
  

  yao-matrix

  and

  ydshieh

  authored
  3bd1c20

  Copy full SHA for 3bd1c20

• Qwen2.5-Omni: Update modeling_qwen2_5_omni.py to fix error when loading quantized weights with AutoAWQ. (#38013)

  Show description for dbc4b91
  tiger-of-shawn

  authored
  dbc4b91

  Copy full SHA for dbc4b91

Commits on May 19, 2025

• Feat: save_pretrained for tensor parallel (and other parallelisms) models (#37919)

  Show description for 46a4b7c
  S1ro1

  authored
  46a4b7c

  Copy full SHA for 46a4b7c

• [doc] fix bugs in `how_to_hack_models.md` (#38198)

  Show description for 9ecee14
  faaany

  authored
  9ecee14

  Copy full SHA for 9ecee14

• Translating model_doc/bert.md to Chinese (#37806)

  Show description for f524439
  Nanji-Huaji

  authored
  f524439

  Copy full SHA for f524439

• Tensor parallel docs (#38178)

  Show description for 6e73841
  S1ro1

  authored
  6e73841

  Copy full SHA for 6e73841

• 🚨🚨🚨 [pipelines] update defaults in pipelines that can `generate` (#38129)

  Show description for 9c50001
  gante

  authored
  9c50001

  Copy full SHA for 9c50001

• [image-text-to-text pipeline] Accept a chat as a positional arg (#38204)

  Show description for 6f9da76
  gante

  authored
  6f9da76

  Copy full SHA for 6f9da76

• [SAM-HQ] Update names in the docs (#38058)

  Show description for 7c9b0ca
  NielsRogge

  authored
  7c9b0ca

  Copy full SHA for 7c9b0ca

• Remove Deprecated `verbose` arg in LayerWiseDummyScheduler (#38197)

  Show description for 04282a9
  DaizeDong

  authored
  04282a9

  Copy full SHA for 04282a9

• Make HF implementation match original OLMo 2 models for lower precisions (#38131)

  Show description for aef1234
  2015aroras

  authored
  aef1234

  Copy full SHA for aef1234

• [docs] add Audio import (#38195)

  Show description for 9644acb
  faaany

  authored
  9644acb

  Copy full SHA for 9644acb

• [docs] minor fixes in `models.md` (#38193)

  Show description for 7d93f93
  faaany

  authored
  7d93f93

  Copy full SHA for 7d93f93

• Pass `eps` to `Mistral3RMSNorm` (#38026)

  Show description for 47f8578
  

  sergiopaniego

  and

  ArthurZucker

  authored
  47f8578

  Copy full SHA for 47f8578

• Resolve Python logger warnings (#38183)

  Show description for 6c63028
  

  emmanuel-ferdman

  and

  github-actions[bot]

  authored
  6c63028

  Copy full SHA for 6c63028

• Support for transformers explicit filename (#38152)

  Show description for 003deb1
  LysandreJik

  authored
  003deb1

  Copy full SHA for 003deb1

• [generation] Less verbose warnings by default (#38179)

  Show description for dbb9813
  gante

  authored
  dbb9813

  Copy full SHA for dbb9813

• Add adam_kwargs for Apollo Optimizer (#38168)

  Show description for 656e2ea
  DaizeDong

  authored
  656e2ea

  Copy full SHA for 656e2ea

• Refactor `get_XXX_dataloader` from Trainer (#38090)

  Show description for 6bb6821
  yaswanth19

  authored
  6bb6821

  Copy full SHA for 6bb6821

Commits on May 16, 2025

• [tests] remove `test_sdpa_equivalence` (redundant) (#37911)

  Show description for 40a493c
  

  gante

  and

  ydshieh

  authored
  40a493c

  Copy full SHA for 40a493c

• fix bug in distributed loss test (#38166)

  Show description for ea29f61
  techkang

  authored
  ea29f61

  Copy full SHA for ea29f61

• Fix import torchao.prototype.low_bit_optim since torchao v0.11 (#38174)

  Show description for a438949
  

  baptxste

  and

  github-actions[bot]

  authored
  a438949

  Copy full SHA for a438949

• Add args support for fast image processors (#37018)

  Show description for 0ba9556
  yonigozlan

  authored
  0ba9556

  Copy full SHA for 0ba9556

• [ESM] Add flash-attention-2 backend for ESM-2 (#38023)

  Show description for d69945e
  pstjohn

  authored
  d69945e

  Copy full SHA for d69945e

• Feat: add warnings for unused keys and rules in tensor parallel (#37893)

  Show description for 7b5e327
  S1ro1

  authored
  7b5e327

  Copy full SHA for 7b5e327

• remove some commands from `fetch_tests` CircleCI job (#38176)

  Show description for 1209352
  ydshieh

  authored
  1209352

  Copy full SHA for 1209352

• Disable `convert to draft` workflow (#38177)

  Show description for 91f6fa0
  ydshieh

  authored
  91f6fa0

  Copy full SHA for 91f6fa0

• Disable `Trigger CircleCI by ready for review` (#38171)

  Show description for 5036ec8
  ydshieh

  authored
  5036ec8

  Copy full SHA for 5036ec8

• clean autoawq cases on xpu (#38163)

  Show description for 7f28da2
  yao-matrix

  authored
  7f28da2

  Copy full SHA for 7f28da2

Pagination

Previous

Next