huggingface/transformers

Pull Requests Commits

MekkCyber committed 162 days ago

a9790b2e

MekkCyber committed 162 days ago

0f4f4f56

Add Youtu-LLM model (#43166)

LuJunru committed 162 days ago

Verified be875640

Fix tiles mixing for batched input, add tie_word_embeddings to LFM2VL config (#43379)

ankke committed 162 days ago

Verified 2a5ba8b5

fix: return labels instead of label in reduce_label method in BeitImageProcessorFast (#43527)

sbucaille committed 163 days ago

Verified be0115e4

[`RoPE`] Make explicit inheritance (#43410)

vasqu committed 163 days ago

Verified 3e27b847

Fix for #43530 (#43535)

Rocketknight1 committed 163 days ago

Verified df58b65b

Operation Green CI (#43530)

Rocketknight1 committed 163 days ago

Verified c3d81076

Tie the weights even if initializing from a config on meta device (#43523)

Cyrilvallez committed 163 days ago

Verified 00f886a9

[kernels] Update cv_utils name (#43529)

MekkCyber committed 163 days ago

Verified 3f254fbf

add trackio to training notebooks (#43442)

merveenoyan committed 163 days ago

Verified 83163c6c

Mark test_prompt_lookup_decoding as flaky (#42184)

Rocketknight1 committed 163 days ago

Verified 14eed387

Fix some MoE routers (#43445)

IlyasMoutawwakil committed 163 days ago

Verified 0dbb56e3

batched_mm is slow on cpu (#43438)

IlyasMoutawwakil committed 163 days ago

Verified a99a913e

fix: initialize BatchNorm2d buffers only when needed (#43520)

tarekziade committed 163 days ago

Verified 0780b0b8

[GLM-OCR] GLM-OCR Support (#43391)

zRzRzRzRzRzRzR committed 163 days ago

Verified 4854dbf9

Fix loading of Qwen3 FP8 (#43494)

githubnemo committed 163 days ago

Verified a1f63d55

fix `ShieldGemma2IntegrationTest::test_model` (#43343)

sywangyi committed 163 days ago

Verified 2d5a6d03

Update `SamHQModelIntegrationTest::test_inference_mask_generation_batched_points_batched_images` for `XPU` (#43511)

sywangyi committed 163 days ago

Verified 461cfb14

Revert utils files changes from PR #42845 (#43507)

ydshieh committed 164 days ago

Verified dfe30827

Move hardcoded time_step params to config for Bamba, FalconH1, GraniteMoeHybrid (#43461)

raimbekovm committed 164 days ago

Verified fa5dfe06

Prepare inputs for generation is called from `super()` (#43280)

zucchini-nlp committed 164 days ago

Verified 3cbf8471

Enhance repo. consistency bot (#43503)

ydshieh committed 164 days ago

Verified 0189e042

Add `pytest-random-order` for reproducible test randomization (#43483)

tarekziade committed 164 days ago

Verified 9495ae28

Add missing GPURawMetrics.from_dict() method in benchmark_v2 (#43499)

Abdennacer-Badaoui committed 164 days ago

Verified c4b4196e

push dev version 5.0.1.dev0

ArthurZucker committed 164 days ago

6301b503

Fix failing `markuplm` & `perception_lm` integration tests (#43464)

Sai-Suraj-27 committed 164 days ago

Verified 60b9abc8

fix(Phi4Multimodal): Fix incorrect default vision/audio config initialization in Phi4MultimodalConfig (#43480)

charlieJ107 committed 164 days ago

Verified 3abe00a9

handle 1D position_ids for modeling_flash_attention_utils as well (#43403)

kaixuanliu committed 164 days ago

Verified 760946c7

Remove stale TODO comments in UDOP tied weights (#43477)

raimbekovm committed 164 days ago

Verified ab87f244

Older