huggingface/transformers

Pull Requests Commits

eustlb committed 4 days ago

583005a8

[gemma3n] fix audio test fixture: use hf_hub_download instead of load_dataset (#49128)

ydshieh committed 5 days ago

Verified 07338b6c

Fix exporters import on torch < 2.9 (is_contiguous_or_false) (#49124)

ydshieh committed 6 days ago

Verified 96331a9f

Remap the legacy Gemma 1 hidden_act in the config post-init (#49084)

PCfVW committed 7 days ago

Verified 27166ea0

[serge] Fix 2 integration tests for model `jamba` failing with `other` (other (2)) (#49044)

sergereview[bot] committed 7 days ago

Verified 6b14ee96

Use grouped_mm on TPU devices under torch.compile (#49097)

salkan0 committed 7 days ago

Verified 6e4bcc5d

Keep the eos ids the config declares (#49082)

albertvillanova committed 7 days ago

Verified e825bfcc

[serge] Fix 2 integration tests for model `pvt_v2` failing with `other` (#49067)

sergereview[bot] committed 7 days ago

Verified f8770a13

[serge] Fix 2 integration tests for model `hy_v3` failing with `output_mismatch` (tensor values differ (2)) (#49068)

sergereview[bot] committed 7 days ago

Verified 4ad04d46

[serge] Fix 2 integration tests for model `cvt` failing with `output_mismatch` (tensor values differ (2)) (#49076)

sergereview[bot] committed 7 days ago

Verified 11c16613

[NemotronH-Omni] Fix device mismatch in test tensor creation (#49102)

ydshieh committed 7 days ago

Verified 5e4d6304

Fix MaskFormerSwin attention mask dtype to follow hidden states (#49032)

kaixuanliu committed 7 days ago

Verified 72c3b930

Use namespaced dataset ids in docs and PyTorch examples (#49017)

davanstrien committed 7 days ago

Verified f0e39b8b

Fix RGB early-return skipping PNG tRNS compositing (#49005)

cs-fisha committed 7 days ago

Verified 6da3313a

[Zamba] Fix associative scan breaking ONNX export and OOM in integration test (#49092)

ydshieh committed 7 days ago

Verified 047de769

[Executorch] Add MLX recipe (#48910)

metascroy committed 7 days ago

Verified 5880561a

Add image processing tester init (#48829)

guarin committed 8 days ago

Verified 89b6b175

Add `Trainer.end` (#48875)

qgallouedec committed 8 days ago

Verified 98d39824

🚨 [ROCm] gpt-oss: route FA3 to aiter-flash-attn, generate ROCm fixtures (#46837)

Abdennacer-Badaoui committed 8 days ago

Verified 763a150e

Keep `attention_mask` as `None` in OPT's causal mask creation (#49002)

jiqing-feng committed 8 days ago

Verified 83445dd7

Bump huggingface_hub upper bound to <3.0 (transformers) (#49083)

Wauplin committed 8 days ago

Verified ff4612bc

Fix NaN in Parakeet eager attention with padded batches (#49070)

ArthurZucker committed 8 days ago

Verified 93b45aa9

Keep already-decoded array arguments unchanged (#49062)

yonigozlan committed 8 days ago

Verified 6d43ab40

Fix odd head_dim validation for RoPE configurations (#48524)

somuai committed 8 days ago

Verified e37e548c

[serge] Fix OOM in Moshi integration tests with MemoryCleanupMixin (#48839)

sergereview[bot] committed 8 days ago

Verified c320b449

Fix assisted eos token condition (#49075)

Cyrilvallez committed 8 days ago

Verified 010f0ca2

Fix XPU Docker image build (#49073)

regisss committed 8 days ago

Verified c47c6abf

Add support for Nemotron Omni (#46509)

meatybobby committed 8 days ago

Verified 4b28d51d

Honor `config.output_router_logits` in the MoE VLM wrappers (#48885)

qgallouedec committed 9 days ago

Verified f3247073

fix(ci): harden GitHub Actions workflows (#49057) (#49059)

hf-security-analysis[bot] committed 9 days ago

Verified 20eeb6ec

Older