huggingface/transformers

Pull Requests Commits

Cyrilvallez committed 12 hours ago

Verified c2fde0ab

reapply all those modualrs

Cyrilvallez committed 12 hours ago

Verified b7922663

remove them from all modulars of llama descendants

Cyrilvallez committed 12 hours ago

Verified f5dc06cd

remove it from llama + modulars

Cyrilvallez committed 12 hours ago

Verified 88a0b2ed

fully remove from cache api

Cyrilvallez committed 12 hours ago

Verified b9672ff4

fully remove from the mask api

Cyrilvallez committed 12 hours ago

Verified 57b58709

[generate] Completely stop relying on `cache_position` to prepare inputs (#44130)

Cyrilvallez committed 12 hours ago

Verified ecf79eb2

[docs] Add Chinese translations for common NLP task tutorials (#44144)

TinderZ committed 14 hours ago

Verified 708d3e12

[`Mamba`] Fix kernel loading (#44176)

vasqu committed 14 hours ago

Verified 87954470

[Mimi] Calibrate to ensure encoder streaming performs correctly (#43971)

caffeinism committed 14 hours ago

Verified 7c0a3be0

ESM2 attention_mask and token_dropout fix (#44163)

lhallee committed 15 hours ago

Verified 9887ca2a

bring back our demons: clean_up_tokenization_spaces (#44035)

ArthurZucker committed 16 hours ago

Verified d2339709

Fix `Seq2SeqTrainingArguments` documentation (#35258)

qgallouedec committed 17 hours ago

Verified 32deb4c3

[`Flash Attn`] Enable compatible implementations (#44177)

vasqu committed 18 hours ago

Verified 8151000f

AutoGrad support for grouped_mm fallback (#44152)

IlyasMoutawwakil committed 19 hours ago

Verified 9a7d8178

Patch `__setitem__` on `ModelOutput` even if the parameter was previously `None` (#44080)

tomaarsen committed 20 hours ago

Verified 00cc937c

[`simple`] Fix up `__repr__` whitespace/brackets (#44048)

tomaarsen committed 20 hours ago

Verified e0b07d1b

[`chore`] Fix incorrect forward type hint for Gemma3n (#44051)

tomaarsen committed 21 hours ago

Verified 64230197

Raise informative error when loading video processors (#44125)

zucchini-nlp committed 22 hours ago

Verified 1f468a8b

fix(flaky): Different approach to make sure loss exists (#43804)

tarekziade committed 23 hours ago

Verified 6b55285d

Add Higgs Audio V2 Model (#40294)

szhengac committed 1 day ago

Verified 1618d44b

[voxtral] fix voxtral proc (#44132)

eustlb committed 1 day ago

Verified 1e31876d

[docs] Fix typos in GenerationConfig docstring (#44143)

nightcityblade committed 1 day ago

Verified 35324377

Fix gemma3n `get_audio_features` (#44040)

zucchini-nlp committed 1 day ago

Verified 4aa550b4

Fix UMT5EncoderModel embedding weights not being tied after loading (#43880)

jiqing-feng committed 1 day ago

Verified a6ef2a6f

fix(testing): Update stale device override test in GraniteSpeech (#44113)

harshaljanjani committed 1 day ago

Verified b240061e

[Misc][vlms] Use text_config when initializing the fine-grained FP8Expert (#44032)

JJJYmmm committed 1 day ago

Verified 196216c3

docs: fix typo 'AuoQuant' → 'AutoQuant' and clarify FINEGRAINED_FP8 library column (#44131)

cluster2600 committed 2 days ago

Verified 556312cd

Update post proc (#44090)

itazap committed 2 days ago

Verified 549835e5

Fix: flaky `Kosmos2ModelTest` test (#44061)

tarekziade committed 2 days ago

Verified 6976f757

Older