huggingface/transformers

Pull Requests Commits

pkooij committed 84 days ago

Verified dcddb970

add openpi changes

pkooij committed 84 days ago

Verified 3676fc80

kashif committed 155 days ago

a5923d4d

refactor: remove `set_tracer_provider` and `set_meter_provider` calls (#39422)

McPatate committed 155 days ago

d04a9425

Cyrilvallez committed 166 days ago

Verified 37f8b0b5

finally fix glm4v accordingly

Cyrilvallez committed 166 days ago

Verified d15e157d

fix attr post conflict

Cyrilvallez committed 166 days ago

Verified 9873b2f9

Cyrilvallez committed 166 days ago

Verified d6f31553

Add a default value for `position_ids` in masking_utils (#39310)

Cyrilvallez committed 166 days ago

Verified cbac4263

fix Glm4v batch videos forward (#39172)

Kuangdd01 committed 166 days ago

Verified c1221458

[sliding window] revert and deprecate (#39301)

zucchini-nlp committed 166 days ago

Verified 9da3f7d0

[smollm3] add tokenizer mapping for `smollm3` (#39271)

gante committed 166 days ago

Verified b2fbc022

[pagged-attention] fix off-by-1 error in pagged attention generation (#39258)

kashif committed 166 days ago

Verified c30a6b7b

Fix errors when use verl to train GLM4.1v model (#39199)

kaln27 committed 166 days ago

Verified be7d1a9d

[bugfix] fix flash attention 2 unavailable error on Ascend NPU (#39166)

FightingZhen committed 166 days ago

Verified 6023ca8a

Fix some bug for finetune and batch infer For GLM-4.1V (#39090)

zRzRzRzRzRzRzR committed 166 days ago

Verified e7e78b2a

Release: v4.53.2

Cyrilvallez committed 166 days ago

Verified 4a5d5c49

Release: v4.53.1

Cyrilvallez committed 173 days ago

Verified 896e9cea

Add packed tensor format support for flex/sdpa/eager through the mask! (#39194)

Cyrilvallez committed 173 days ago

Verified bc0a47e6

when delaying optimizer creation only prepare the model (#39152)

winglian committed 173 days ago

Verified 63af3d7a

Fix multimodal processor get duplicate arguments when receive kwargs for initialization (#39125)

Isotr0py committed 173 days ago

Verified a10c2bf8

[smolvlm] fix video inference (#39147)

zucchini-nlp committed 173 days ago

Verified 650bc679

[qwen2-vl] fix FA2 inference (#39121)

zucchini-nlp committed 173 days ago

Verified 3c5f9108

Several fixes for Gemma3n (#39135)

Cyrilvallez committed 173 days ago

Verified 5e1c9146

Fix key mapping for VLMs (#39029)

bvantuan committed 173 days ago

Verified 8446e2a5

Fix: unprotected import of tp plugin (#39083)

S1ro1 committed 173 days ago

Verified 6900fe8a

Release: v4.53.0

LysandreJik committed 181 days ago

67ddc82f

[Modeling] Fix encoder CPU offloading for whisper (#38994)

kylesayrs committed 181 days ago

Verified 0a8081b0

Gemma 3n (#39059)

RyanMullins committed 181 days ago

Verified c63cfd6a

[tests] remove tests from libraries with deprecated support (flax, tensorflow_text, ...) (#39051)

gante committed 181 days ago

Verified 3e5cc128

Older