huggingface/transformers

Pull Requests Commits

molbap committed 42 days ago

d589c977

call transformers-ci in a nightly run (#46811)

tarekziade committed 43 days ago

Verified 7b89511f

[docs] full disk offloading (#46893)

stevhliu committed 43 days ago

Verified c96378c4

TST Run fast PEFT tests in normal CI (#45679)

BenjaminBossan committed 43 days ago

Verified 2cae6474

🚨 EP: fix EP router contract for many models + honor FP8 scale format (#46818)

IlyasMoutawwakil committed 43 days ago

Verified 8edd87b6

[docs] AITER kernels (#46871)

stevhliu committed 43 days ago

Verified db7c3720

nemotron_asr_streaming: set _supports_flex_attn to False (#46878)

kaixuanliu committed 43 days ago

Verified dc5a497c

Add native masked MSE loss for Sapiens2ForPoseEstimation (#46764)

Sainava committed 43 days ago

Verified 33d1bc04

blip 2 fix (#46816)

itazap committed 43 days ago

Verified 08428b23

Use meshgrid for brevity (#46861)

zucchini-nlp committed 43 days ago

Verified cf284d4f

Add xcodec2 model (#44178)

ebezzam committed 43 days ago

Verified 9b6af5d7

Rework some old cache getters/properties (#46862)

Cyrilvallez committed 44 days ago

Verified 587a429a

Prevent auto-class from being modified for all models (#46844)

zucchini-nlp committed 44 days ago

Verified c21da1b5

Add Spanish translation of the torch.compile page (#46852)

delcenjo committed 44 days ago

Verified 353b1532

docs: Update NeMo AutoModel doc examples (#46857)

adil-a committed 44 days ago

Verified 3f21c47f

[docs] distributed training (#44420)

stevhliu committed 44 days ago

Verified 3064b654

[docs] require trust_remote_code for custom_generate (#46677)

stevhliu committed 44 days ago

Verified f73cc1b1

Documentation for the kernel API (#46754)

michaelbenayoun committed 44 days ago

Verified 50b4cc18

add distributed config (#46705)

3outeille committed 44 days ago

Verified 2e2855f0

[Offloading] [Bugfix] Fix disk offloading of models with explicit tensor dtypes (#46849)

kylesayrs committed 44 days ago

Verified 176a17bd

Streamable chat parsing (#45847)

Rocketknight1 committed 44 days ago

Verified 4ab0479e

Round the ue8m0 FP8 scale before quantizing so dequant matches the stored inverse (#46763)

Incheonkirin committed 44 days ago

Verified e32afc72

Fix BitNet packed-weight unpacking dtype (`F.linear` dtype mismatch) (#46808)

jiqing-feng committed 44 days ago

Verified 22d94091

Fix typos in code (#46579)

cyyever committed 44 days ago

Verified 19d179f2

Fix Evolla eager attention for the GQA text decoder (#46860)

jiqing-feng committed 44 days ago

Verified 6b1baa93

Fix Moonshine training-loss double-shift (train against labels, not labels[..., 1:]) (#46784)

Incheonkirin committed 44 days ago

Verified d8c23549

[CB] Fix issues with FA read / writes (#46765)

remi-or committed 45 days ago

Verified d87f670d

Fix expanded dim in the cache's cumulative length (#46856)

Cyrilvallez committed 45 days ago

Verified 8dbac377

Switch decorator order (#46853)

Cyrilvallez committed 45 days ago

Verified 32ec18bb

Fix beam search for mamba models (#46819)

Cyrilvallez committed 45 days ago

Verified 42d7b005

Older