huggingface/transformers

Pull Requests Commits

Switch gradient checkpointing default to `use_reentrant=False` (PyTorch recommended) (#43203)

qgallouedec committed 31 days ago

Verified 625ad844

[mimi] mimi conv cache edge case (#43130)

eustlb committed 31 days ago

Verified 6d4d9841

Untangle config inheritance (#41541)

zucchini-nlp committed 31 days ago

Verified aec03e5d

Do not skip integration tests (#42899)

zucchini-nlp committed 31 days ago

Verified 79f03863

add comment explaining limitation

IlyasMoutawwakil committed 31 days ago

f094c319

[`FA`] Generalize fa config checks and fix flags (#43121)

vasqu committed 32 days ago

Verified 9eea1b0b

[doc] glm-asr is an audio model (#43309)

eustlb committed 32 days ago

Verified 847a68df

[LWDETR] update model repos from stevenbucaille to AnnaZhang (#43266)

sbucaille committed 32 days ago

Verified d9c61949

Fix eager experts for DeepSeekV2 (#43288)

Rocketknight1 committed 32 days ago

Verified 396caf56

Fix MusicGen generation config (#43194)

zucchini-nlp committed 32 days ago

Verified 38e5987e

Merge branch 'main' into gpt-oss-experts-impl

IlyasMoutawwakil committed 32 days ago

Verified aa34996f

Simplify `_get_resolved_checkpoint_files` and do not try safetensors bot conversion with a subfolder (#43301)

Cyrilvallez committed 32 days ago

Verified d12672fc

no need for logger

IlyasMoutawwakil committed 32 days ago

4b0323ce

avoid baddbmm as it is less performant / less optimizable by max-autotune

IlyasMoutawwakil committed 32 days ago

1c491124

quantization conversions should be applied first

IlyasMoutawwakil committed 32 days ago

52e07786

IlyasMoutawwakil committed 32 days ago

dd377e19

fix subfolder issue (#43300)

ArthurZucker committed 32 days ago

Verified a85d6b2a

force 16 bytes alignmenet during weight loading

IlyasMoutawwakil committed 32 days ago

2f3fd11c

[Qwen-VL] min/max pixels gone forever (#43279)

zucchini-nlp committed 32 days ago

Verified 2b13b08d

IlyasMoutawwakil committed 32 days ago

9dbed89b

[docs] remove old docs (#43156)

stevhliu committed 33 days ago

Verified 7be2d358

Fix pipeline tests (#43293)

Cyrilvallez committed 33 days ago

Verified 405d18df

[CB] Refactor: extract common scheduling logic into Scheduler base class (#43237)

dstekanov committed 33 days ago

Verified 3af7c35f

Fix failing `fast_vlm` tests (#43275)

Sai-Suraj-27 committed 33 days ago

Verified 0a5d574d

Delete Text2TextPipeline, SummarizationPipeline and TranslationPipeline (#43256)

Rocketknight1 committed 33 days ago

Verified fc532243

[`Docs`] Minimax M2 add release date (#43290)

vasqu committed 33 days ago

Verified 34691a98

[ci] Fixing some failing tests for important models (#43231)

Abdennacer-Badaoui committed 33 days ago

Verified d1808f2c

Add LightOnOCR model implementation (#41621)

baptiste-aubertin committed 33 days ago

Verified 3d276453

Update src/transformers/models/gpt_oss/modeling_gpt_oss.py

IlyasMoutawwakil committed 33 days ago

Verified b1a71a79

fix crash in when running FSDP2+TP (#43226)

sywangyi committed 33 days ago

Verified 77146cc9

Newer Older