NielsRogge/transformers

Pull Requests Commits

fix(testing): Fix LayoutXLM tokenization test and LightOnOCR SDPA flash test failures on main CI (#43988)

harshaljanjani committed 15 days ago

Verified 65935e09

Add ModernVBERT models (#42504)

paultltc committed 15 days ago

Verified 9f43e7b4

fix: HiggsAudioV2 cached decode inputs in compiled generation (#44201)

tarekziade committed 15 days ago

Verified 78c873ef

fix(models): Fix LayoutLMv2 NER crash and broken batched truncation/padding (#44187)

harshaljanjani committed 16 days ago

Verified a3dcad9e

[CB] [Major] Asynchronous batching (#43960)

remi-or committed 16 days ago

Verified 9f7faed2

Fix LASR feature extractor regression from invalid center argument (#44207)

ainergiz committed 16 days ago

Verified 5ef42f53

Models with incorrect tokenizer_class in tokenization_config.json tha… (#44179)

itazap committed 16 days ago

Verified df1cd3a7

chore(typing): initial ty integration (#44167)

tarekziade committed 18 days ago

Verified 147b7aa0

fix(flaky): `test_generate_with_and_without_position_ids` in GLM ORC (#44173)

tarekziade committed 18 days ago

Verified a3bfc8bc

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

Cyrilvallez committed 18 days ago

Verified ecf79eb2

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

TinderZ committed 18 days ago

Verified 708d3e12

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

vasqu committed 18 days ago

Verified 87954470

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

caffeinism committed 18 days ago

Verified 7c0a3be0

ESM2 attention_mask and token_dropout fix (#44163)

lhallee committed 18 days ago

Verified 9887ca2a

bring back our demons: clean_up_tokenization_spaces (#44035)

ArthurZucker committed 18 days ago

Verified d2339709

Fix `Seq2SeqTrainingArguments` documentation (#35258)

qgallouedec committed 18 days ago

Verified 32deb4c3

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

vasqu committed 18 days ago

Verified 8151000f

AutoGrad support for grouped_mm fallback (#44152)

IlyasMoutawwakil committed 18 days ago

Verified 9a7d8178

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

tomaarsen committed 19 days ago

Verified 00cc937c

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

tomaarsen committed 19 days ago

Verified e0b07d1b

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

tomaarsen committed 19 days ago

Verified 64230197

Raise informative error when loading video processors (#44125)

zucchini-nlp committed 19 days ago

Verified 1f468a8b

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

tarekziade committed 19 days ago

Verified 6b55285d

Add Higgs Audio V2 Model (#40294)

szhengac committed 19 days ago

Verified 1618d44b

[voxtral] fix voxtral proc (#44132)

eustlb committed 19 days ago

Verified 1e31876d

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

nightcityblade committed 19 days ago

Verified 35324377

Fix gemma3n `get_audio_features` (#44040)

zucchini-nlp committed 19 days ago

Verified 4aa550b4

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

jiqing-feng committed 19 days ago

Verified a6ef2a6f

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

harshaljanjani committed 19 days ago

Verified b240061e

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

JJJYmmm committed 20 days ago

Verified 196216c3

Older