huggingface/transformers

Pull Requests Commits

remaining modalities

SunMarc committed 3 hours ago

471cf880

stricker check for now

SunMarc committed 4 hours ago

f755bd26

revert some changes that shouldn't be done

SunMarc committed 4 hours ago

f14af2e2

SunMarc committed 4 hours ago

8679334b

SunMarc committed 4 hours ago

dc604d33

Merge main into moe-aux-loss-scaling

SunMarc committed 9 hours ago

ede58608

OpenVINO exporter: let OpenVINO fuse attention, RoPE, the KV cache and compressed matmuls (#50311)

IlyasMoutawwakil committed 10 hours ago

Verified 72b487d1

[VibeVoice] Avoid GPU syncs when removing audio padding (#50417)

hmellor committed 11 hours ago

Verified 9fa60240

[vLLM] Add Omni token counting and text replacement offsets for the Transformers backend (#49435)

harshaljanjani committed 14 hours ago

Verified 6df64031

Subconfig creation in testers (#49528)

zucchini-nlp committed 16 hours ago

Verified 85db0e06

[`distributed`] Trainer: reduce FSDP/HSDP gradients on the last micro-batch only (#50286)

3outeille committed 17 hours ago

Verified 8833c918

[`distributed`] Support HSDP with expert parallelism (#50293)

3outeille committed 17 hours ago

Verified bbe484b5

[`distributed`] Add HSDP (`dp_replicate_size`) to `DistributedConfig` (#50285)

3outeille committed 18 hours ago

Verified 522cbe68

ZeRO-3: read weights on rank 0 only and convert them one group at a time (#50284)

3outeille committed 18 hours ago

Verified 677b9ff1

Multmodal token count validation in processors (#49976)

zucchini-nlp committed 18 hours ago

Verified fa421d61

Remove deprecated refs (#50374)

zucchini-nlp committed 19 hours ago

Verified f262c717

Fix Mistral Large 4 config (#50418)

Cyrilvallez committed 1 day ago

Verified cdb75de4

[docs] Common quantization tasks (#50095)

stevhliu committed 1 day ago

Verified 7d5d6d66

Fix flaky `Gemma4Audio2TextModelTest::test_training_gradient_checkpointing_use_reentrant_true` (#50410)

ydshieh committed 1 day ago

Verified 084b1fd1

Fix flaky `T5GemmaModelTest::test_decoder_model_past_with_large_inputs` (#50409)

ydshieh committed 1 day ago

Verified 7b0fec70

[Model] LFM3 Family: LFM3-Text (#50322)

paulpak58 committed 1 day ago

Verified 7d6cae08

Don't pick torchcodec when it can't load or has no FFmpeg (#50362)

eustlb committed 1 day ago

Verified 5ea23808

[CB] Disable streams on MPS to fix continuous batching with torch 2.15 (#50404)

Isalia20 committed 1 day ago

Verified 2ffebc8c

Keep enabling checkpointing on models that only define gradient_checkpointing_enable (#50349)

qgallouedec committed 1 day ago

Verified a15175e3

Fix flaky `Owlv2ModelTest::test_eager_matches_sdpa_inference_01_fp16_pad_left` (#50403)

ydshieh committed 1 day ago

Verified f1eff635

SunMarc committed 1 day ago

7826cc76

Fix flaky `PocketTTSAudioTokenizerModelTest::test_batching_equivalence` on CPU (#50402)

ydshieh committed 1 day ago

Verified 9bb266d5

[`SDPA`] Fix decode under compile (#50388)

vasqu committed 1 day ago

Verified 2aab563b

fix(pipelines): validate zero-shot detection batch labels (#50318)

BlazeSR10hawk committed 1 day ago

Verified b55e0761

Fix PaddleOCR-VL vision position interpolation alignment (#50384)

scyyh11 committed 1 day ago

Verified 924266c4

Older