huggingface/transformers

Pull Requests Commits

Rocketknight1 committed 7 hours ago

aaf1e042

Rocketknight1 committed 7 hours ago

5ffe6ed8

Reduce peak memory in examples_torch CI job (OOM fix) (#48983)

ydshieh committed 8 hours ago

Verified 21912efc

Fix typos in DeepseekV4 comments (#48953)

Janiarafath committed 9 hours ago

Verified ae05b6b9

Contract the mamba2 chunk scan with einsum instead of broadcast-then-sum (#48978)

tarekziade committed 9 hours ago

Verified a52f6593

Fix image processor class-level size mutation and min_pixels handling (#48916)

Gracy769 committed 10 hours ago

Verified d53e5876

Add pose estimation keypoint preprocessing to Sapiens2ImageProcessor (#47199)

Sainava committed 10 hours ago

Verified e86881ed

Better attn default for gguf (#48935)

SunMarc committed 12 hours ago

Verified abd76d12

Skip an unnecessary image copy in the torchvision image normalization path (#48897)

jzakrzew committed 12 hours ago

Verified 207fca7b

Fix more stuff (#48979)

zucchini-nlp committed 14 hours ago

Verified c210bb39

Fix infeasible cost matrix errors in hugarian matcher losses (#47730)

guarin committed 14 hours ago

Verified e138c4d5

Keep image processor backends in sync on keys and dtypes (#48739)

yupengtang committed 15 hours ago

Verified a7d9d949

Resolve the Hub revision once per load instead of passing a private _commit_hash around (#47611)

Wauplin committed 21 hours ago

Verified d67c7293

Fix AXK2 integration test: update CUDA expected text and rename class (#48941)

ydshieh committed 3 days ago

Verified c587bc88

docs: fix dead doc links in i18n READMEs and the xlnet docstring (#48905)

simpleqt committed 3 days ago

Verified 3713bd83

Align special tokens on the text config (#48847)

qgallouedec committed 3 days ago

Verified ea31b0c6

DOC: Improve documentation for remap_legacy_layer_types function (#48651)

mathewOracle committed 3 days ago

Verified 266cc923

docs: fix docstring parameters that do not match signatures (#48904)

simpleqt committed 3 days ago

Verified c9ad4a71

Fix tests due to dropping attn mask (#48903)

SunMarc committed 3 days ago

Verified b856a67b

Improve lazy import error messages (#48602)

karatarassul4-max committed 3 days ago

Verified 7301fbc6

fix(vibevoice-asr): use integer ceiling division for audio token count (#48864)

ege-arhan committed 3 days ago

Verified 719d8090

Modular conversion small fix (#48934)

zucchini-nlp committed 3 days ago

Verified 95ad2ed0

Fix possessive typo in Whisper long-form warning (#48877)

erikdw committed 3 days ago

Verified bdb4cc00

Standardise Aria's MoE onto the experts interface (#48907)

hmellor committed 3 days ago

Verified fd204af8

Update arXiv citation in NeoMME model doc (#48926)

tonywu71 committed 3 days ago

Verified 18556136

Fix compile cache tests (#48923)

Cyrilvallez committed 3 days ago

Verified 9b819e72

Size the device_map buffer from the largest leaf module (#47211)

dhruv7477 committed 3 days ago

Verified 4618eba1

Fix T5 tied weights order for lm_head (#48238)

vaibhavmashal committed 3 days ago

Verified 483b8806

Clean up some MoE models' integration tests (#48833)

ydshieh committed 4 days ago

Verified 6c6bac29

Mark test_generate_with_static_cache as flaky for olmo and bigbird_pegasus (#48856)

ydshieh committed 4 days ago

Verified da58b892

Older