huggingface/transformers

Pull Requests Commits

chore(sec): added a handful of security checks

tarekziade committed 25 days ago

d2815c19

avoid wrap 4bit-quantized model into DP (#45407)

kaixuanliu committed 26 days ago

Verified 168c5b68

Add example for iterative chatting with MLLMs (#45398)

zucchini-nlp committed 26 days ago

Verified 5c77ee2d

Gemma4 resizing per layer inputs (#45324)

zucchini-nlp committed 26 days ago

Verified b6f9463e

Add `step3_vl` to `MODELS_WITH_INCORRECT_HUB_TOKENIZER_CLASS` (#45449)

hmellor committed 26 days ago

Verified 331ea339

Update workflow references to new commit hash (#45442)

paulinebm committed 26 days ago

Verified 7028c30b

[Gemma4] Add docstrings for Per-Layer Embeddings (PLE) pipeline (#45207)

w4nderlust committed 26 days ago

Verified 155db714

[Doc] Correct checkpoint path in Dinov2 model_docs (#45430)

ambroiseodt committed 26 days ago

Verified 99ce466f

Fix ty for transformers cli (#45190)

SunMarc committed 26 days ago

Verified 5e3af512

fix(models): Resolve regressions in Wav2Vec2PhonemeCTCTokenizer (wav2vec2-lv-60-espeak-cv-ft) (#45199)

harshaljanjani committed 26 days ago

Verified 5b565a58

fix: prevent accelerate from splitting vision encoder by setting _no_… (#43047)

CodersAcademy006 committed 26 days ago

Verified 27fbb514

Fix Qwen2.5VL temporal grid positions (#45400)

zucchini-nlp committed 26 days ago

Verified fd63aa22

[`fix`] PEFT integration fixes preventing save/load & integration (#45428)

tomaarsen committed 27 days ago

Verified 4396b1b5

Fix the response schema for the gemma4 converter (#45411)

Rocketknight1 committed 27 days ago

Verified 908064cb

[Doc] MoE routing capture and replay recipe (#44925)

kashif committed 27 days ago

Verified def5e686

Fix `apply_chat_template` crash on `tool_call` messages without content (#45348)

qgallouedec committed 27 days ago

Verified e40b0c01

Add SAM3-LiteText (#44320)

NielsRogge committed 27 days ago

Verified 22278df3

Fix IndexError with DeepSpeed ZeRO-3 when kernels rotary is active (#45414)

ArthurZucker committed 27 days ago

Verified 52f2268b

[AMD CI] Fix torch.compile/export failures on AMD CI due to untraceable set.__contains__ (#45282)

Abdennacer-Badaoui committed 27 days ago

Verified 87a69a3f

[inference_fusion] convert conv3d patch embed to linear (#45041)

JJJYmmm committed 27 days ago

Verified 3a947e28

Fix Kimi-K2.5 tokenizer regression and _patch_mistral_regex AttributeError (#45359)

ArthurZucker committed 27 days ago

Verified 282078b7

fix(serving): resolve rust tokenizer from ProcessorMixin in streaming generation (#45368)

sharziki committed 27 days ago

Verified a1b89d7f

Fix #45305 + add regression test GAS (#45349)

florian6973 committed 27 days ago

Verified 20b6f538

Update `trackio` integration to use Buckets and "freeze" Space after training (#45329)

abidlabs committed 27 days ago

Verified 7afee868

fix(qwen3_moe): correct return type annotation on Qwen3MoeSparseMoeBlock.forward (#45352)

RudrenduPaul committed 27 days ago

Verified 357f414c

Fix: NotebookProgressCallback crash when evaluating with the Trainer (#44949)

Charly21r committed 27 days ago

Verified 0b5dbfc9

docs: fix 5 docstring errors in Gemma3nTextConfig (typos, grammar, formatting) (#45370)

RudrenduPaul committed 27 days ago

Verified 6db69b91

Less unnecessary RoPE warnings (#45289)

zucchini-nlp committed 27 days ago

Verified 19a7c100

[`Tokenizers`] Move gpt sw3 tokenizer out (#45404)

vasqu committed 27 days ago

Verified 2c498dfd

Fix unintended Hub metadata calls from _patch_mistral_regex (#43603)

vaibhav-research committed 28 days ago

Verified def8e6a2

Older