huggingface/transformers

Pull Requests Commits

Fix FSDP loading of concatenated checkpoint chunks

3outeille committed 36 days ago

7d7fb013

:rotating_light: [`Kernels`] Bump version (#48714)

vasqu committed 37 days ago

Verified bcb9849a

[Fix] Clean-up ternaries in the DeepSeek family (#48447)

remi-or committed 37 days ago

Verified 6cd7ae39

Fix decoder only path for older bert variants (#48785)

Cyrilvallez committed 37 days ago

Verified 3aba670a

[docs] mrope and axial rope (#48717)

stevhliu committed 37 days ago

Verified 2cba1950

Fix `AutoModel.from_pretrained` not restoring `modules_to_save` weights (#48595)

shniubobo committed 37 days ago

Verified dbbd5512

Add integration tests for MuseGlimmerAssistantModel (#48796)

ydshieh committed 37 days ago

Verified 39f81b04

Fix kosmos flaky test (#48800)

IlyasMoutawwakil committed 37 days ago

Verified 229cd066

Restore legacy tensor-parallel initialization for compatibility (#48797)

3outeille committed 37 days ago

Verified a1ba1c14

[serge] Fix 12 integration tests for model `edgetam` failing with `import_or_config` (other (12)) (#48322)

sergereview[bot] committed 37 days ago

Verified d72aef09

Bump `peft` version requirement (#48716)

shniubobo committed 37 days ago

Verified 75c3583e

add kernel mapping entry for RMSNormGated, KDA, Conv1D on XPU (#48702)

kaixuanliu committed 37 days ago

Verified ca3bdf7a

Cast pixel values to the patch embedding dtype in DeepSeek-OCR-2 (#48632)

jiqing-feng committed 37 days ago

Verified a20852a0

Fix TextToAudioPipeline crash for tokenizer-only models (#48505)

jiqing-feng committed 37 days ago

Verified 9eb65c32

Fix inkling embedding norm (#48786)

Cyrilvallez committed 37 days ago

Verified 33849085

Fix hidden state selection on Gemma4 assistant's first prefill step (#48704)

glistening committed 38 days ago

Verified ff377ea4

Kernel api doc, part 2 (#46889)

michaelbenayoun committed 38 days ago

Verified 7219d282

[InternVL] Normalize num_patches before np.cumsum (#48469)

lorenzozanee committed 38 days ago

Verified 9ab9957d

Fix Glm4vMoeIntegrationTest: offload_folder + MemoryCleanupMixin (#48776)

ydshieh committed 38 days ago

Verified f24b457b

Fix silently random-initializing `RTDetrModel`/`SEWDForCTC` loads (wrong `base_model_prefix`) (#48744)

abhinav-phi committed 38 days ago

Verified 56c5e876

fix videomae load error (#48675)

sywangyi committed 38 days ago

Verified 06a7a68e

Switch daily CI to torch 2.14 — update expected outputs (#48750)

ydshieh committed 38 days ago

Verified a2c15b30

Enable FSDP2 + expert parallelism via a 2-D (fsdp, tp) device mesh (#48516)

qgallouedec committed 39 days ago

Verified 415e6d2f

unifying device_mesh init to enable PP + TP inference (#48155)

3outeille committed 39 days ago

Verified 5474a55e

Use huggingface_hub httpx export (#48685)

Wauplin committed 40 days ago

Verified 80a79d1b

[DeepseekV3] OOM cascade root-cause investigation (generator ref leak in conversion_mapping) (#48720)

ydshieh committed 40 days ago

Verified df04b012

Keep the expert-parallel sentinel slots out of the router gradient (#48689)

qgallouedec committed 40 days ago

Verified 67f50855

add HyperClovaX Vision (#44314)

jp1924 committed 40 days ago

Verified a6cf156d

Add flash_attention_4 in attn_implementation AutoModel docstring (#48684)

3manifold committed 40 days ago

Verified 9c944e97

[CI] Replace hardcoded username allowlists with author_association check in workflow triggers (#48712)

ydshieh committed 40 days ago

Verified 3f601734

Older