huggingface/transformers

Pull Requests Commits

Relax FA3 decorator so it matches runtime

remi-or committed 22 days ago

5fed3c23

fix(cli/serve): import serve handlers lazily so the CLI works without the `serve` extra (#46473)

Anai-Guo committed 22 days ago

Verified 4d47b061

[`CB`] Add FA2 to the fast path (#46729)

vasqu committed 22 days ago

Verified 43639365

update kernels-community/aiter-rope version (#46810)

Abdennacer-Badaoui committed 23 days ago

Verified 8fea7d39

Fix prompt lookup decoding crash when no EOS token is configured (#46790)

Sunt-ing committed 23 days ago

Verified 123f5dd7

Fix flex_attention block mask creation when `get_seq_length` returns a tensor (#46802)

jiqing-feng committed 23 days ago

Verified f007617e

Fix left-padding token selection in `BioGptForSequenceClassification` (#46782)

Sunt-ing committed 23 days ago

Verified bfd3604d

Fix broken internal links in model documentation (#46807)

ShamSaleem committed 23 days ago

Verified be326dbd

DiffusionGemma: mask layout and CI (#46654)

zucchini-nlp committed 23 days ago

Verified 14917fc5

Use cached added-token dicts in per-token decode loops (#46535)

ishan-1010 committed 23 days ago

Verified 7f61d468

Fix offloaded cache device mismatch on hybrid models (#46748)

Sunt-ing committed 23 days ago

Verified 6446f1fc

fix another flaky test (#46767)

zucchini-nlp committed 23 days ago

Verified 75e051d2

Fix secondary rate limit when downloading artifacts in slack report (#46796)

ydshieh committed 23 days ago

Verified 7bd2725e

Add Videoprism (#39895)

MHRDYN7 committed 26 days ago

Verified 1048e9af

docs: move SmolLM3 to Text models category in _toctree.yml (#46770)

yyouretoast committed 26 days ago

Verified a6f8ab79

Add GB10/SM121 Hub-kernel path for Qwen3.6 Gated DeltaNet (#46423)

AzeezIsh committed 26 days ago

Verified 95ed1b82

Add Nemotron ASR Streaming (#46332)

eustlb committed 26 days ago

Verified ed51a079

Fix several bugs in `cache_implementation=static` (#46446)

dacorvo committed 26 days ago

Verified e7835fba

[CI] Fix artifact download path in self-comment-ci workflow (#46769)

ydshieh committed 26 days ago

Verified c4153521

fixes per head minimaxm3 (#46719)

ArthurZucker committed 26 days ago

Verified 2a4ead18

[fix] enable base64 str audio in load_audio (#46694)

eustlb committed 26 days ago

Verified 4f041363

[`Kernels`] Trigger proper kernelization on `use_kernels=True` (#46755)

vasqu committed 26 days ago

Verified 06432852

Lazily build the default kernel mapping to decouple `kernels` from normal transformers usage (#46681)

jiqing-feng committed 26 days ago

Verified c2c13860

[`CI`] Fix some failures introduced by myself :grimacing: (#46751)

vasqu committed 26 days ago

Verified 354533eb

Fix regression in ProcessorMixin._load_tokenizer_from_pretrained for tokenizers at root (#46592)

punyamodi committed 26 days ago

Verified a53a8d8a

Update tokenizer mappings to use TokenizersBackend for additional models (#46091)

itazap committed 26 days ago

Verified efe40496

fix(aria): use math.ceil in get_number_of_image_patches to match actual patch count (#46732)

arnavkewalram committed 26 days ago

Verified d8904a23

Return logits from semantic segmentation post-process (#46163)

guarin committed 26 days ago

Verified fc5401d1

Fall back to the for-loop grouped_mm on CPU (#46743)

Sunt-ing committed 26 days ago

Verified b4f9cd23

Kernelize refactor (#46520)

michaelbenayoun committed 27 days ago

Verified ad697ec1

Older