huggingface/transformers

Pull Requests Commits

fix conflicts in update

molbap committed 7 days ago

8b9eced8

molbap committed 7 days ago

5f4a1ca7

big update for processing kernels

molbap committed 7 days ago

147671e6

Post two CI badges on a PR: CPU PR CI and GPU run-slow (#48190)

tarekziade committed 7 days ago

Verified e453228e

Fix stale expected values in integration tests (cuda sm_86 / Aug04 regressions) (#48171)

ydshieh committed 7 days ago

Verified 3ad85ca4

Assign a reviewer even when a codeowner has left, and route models by modality (#48085)

tarekziade committed 7 days ago

Verified e277d1ed

[VITS] Un-skip test_model_forward (#46375)

blipbyte committed 7 days ago

Verified bb593bfb

Apply context parallelism to the evaluation path (#48167)

qgallouedec committed 7 days ago

Verified 695bd682

Fix `gpt_oss` runs on GPU (#48118)

tarekziade committed 7 days ago

Verified 275b5245

🚨 TP dtensor API inference + training (#47579)

3outeille committed 7 days ago

Verified 861f4c41

[EsmFold2] Fix stale expected distogram logit values (#48182)

ydshieh committed 7 days ago

Verified d4c297d1

Fix Apr 05 integration test regressions (cuda sm_86) (#48170)

ydshieh committed 7 days ago

Verified cc33c370

Add MLU support to is_flash_linear_attention_available (#46995)

atri2549 committed 7 days ago

Verified 6a9e2d37

Fix ROCm SDPA-flash skip guard that crashes on RDNA GPUs (#47965)

Abdennacer-Badaoui committed 7 days ago

Verified 3cb4f80f

Fix integration test expected values for cuda sm_86 (Mar 15 regressions) (#48168)

ydshieh committed 7 days ago

Verified 251c2d4f

Fix DynamicCache reconstruction during ExecuTorch export (#47900)

eladsegal committed 7 days ago

Verified 3cd1a230

Revert "Support per-layer cache configuration and attention-mask selection" (#48175)

Cyrilvallez committed 7 days ago

Verified 6b3a9de1

Support per-layer cache configuration and attention-mask selection (#47901)

eladsegal committed 7 days ago

Verified c84ac341

[LLaVA] Fix pixtral integration tests for cuda sm_86 (#48166)

ydshieh committed 8 days ago

Verified 945dac91

Port ESMC and ESMFold2 to Transformers (#46419)

Rocketknight1 committed 8 days ago

Verified c8df3fc9

[Qwen2.5-Omni] Update stale expected values for cuda sm_86 (#48164)

ydshieh committed 8 days ago

Verified cdea840f

[Mistral3] Fix batched integration tests: padding_side=left + update expected values (#48161)

ydshieh committed 8 days ago

Verified dd6c64b7

Fix DeepSeek V2 default vocab size (#48159)

hmellor committed 8 days ago

Verified e4dd56a3

[InternVL] Fix stale expected values for Llama integration tests (cuda sm_80) (#48153)

ydshieh committed 8 days ago

Verified 0c928118

Retry transient network errors (RemoteDisconnected) in github_utils (#48124)

ydshieh committed 8 days ago

Verified 95119511

Revert "[Quantization]: Refactor is_quantization_compressed for format-based detection" (#48072)

subin9 committed 8 days ago

Verified 238a2dbd

[Whisper] Fix speculative decoding: preserve cleared suppress tokens through super().generate() (#48108)

ydshieh committed 8 days ago

Verified b6688043

🚨[wav2vec2] Support attn_implementation=sdpa dispatch (#46196)

YangKai0616 committed 8 days ago

Verified 2b296305

fix bugs for clvp model (#47127)

kaixuanliu committed 8 days ago

Verified 9ad4b858

Always tie embeddings for LongT5 and Pop2Piano (#47620)

jiqing-feng committed 8 days ago

Verified 5909a467

Older