huggingface/transformers

Pull Requests Commits

Increase timeout when preparing CI (#43560)

Rocketknight1 committed 264 days ago

Verified 5752665b

43054: Add Siglip2Tokenizer to enforce training-time text preprocessing defaults (#43101)

vaibhav-research committed 264 days ago

Verified 1be63d94

check PR bot permission - part 3 (try content attribute) (#43555)

ydshieh committed 264 days ago

Verified a63032b6

check PR bot permission - part 2 (style only) (#43554)

ydshieh committed 264 days ago

Verified 0b563e16

check PR bot permission - part 1 (#43553)

ydshieh committed 264 days ago

Verified 5b584285

Fix failing tests due to no attribute `pad_token_id` (#43453)

Sai-Suraj-27 committed 264 days ago

Verified 28646009

fix: GPT OSS Conversion Script Enhancements (#42901)

KyleMylonakisProtopia committed 264 days ago

Verified cfe89b27

[Quantization] Fix triton_kernels name after being renamed to gpt-oss-triton-kernels (#43528)

MekkCyber committed 264 days ago

Verified 9238fc3d

[Quantization] Add cutlass kernel for FP8 (#43304)

MekkCyber committed 264 days ago

Verified 2b7bc596

[CB] Minor perf improvements and ty compatibility (#43521)

remi-or committed 264 days ago

Verified b207f38e

Add Youtu-LLM model (#43166)

LuJunru committed 264 days ago

Verified be875640

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

ankke committed 264 days ago

Verified 2a5ba8b5

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

sbucaille committed 264 days ago

Verified be0115e4

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

vasqu committed 265 days ago

Verified 3e27b847

Fix for #43530 (#43535)

Rocketknight1 committed 265 days ago

Verified df58b65b

Operation Green CI (#43530)

Rocketknight1 committed 265 days ago

Verified c3d81076

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

Cyrilvallez committed 265 days ago

Verified 00f886a9

[kernels] Update cv_utils name (#43529)

MekkCyber committed 265 days ago

Verified 3f254fbf

add trackio to training notebooks (#43442)

merveenoyan committed 265 days ago

Verified 83163c6c

Mark test_prompt_lookup_decoding as flaky (#42184)

Rocketknight1 committed 265 days ago

Verified 14eed387

Fix some MoE routers (#43445)

IlyasMoutawwakil committed 265 days ago

Verified 0dbb56e3

batched_mm is slow on cpu (#43438)

IlyasMoutawwakil committed 265 days ago

Verified a99a913e

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

tarekziade committed 265 days ago

Verified 0780b0b8

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

zRzRzRzRzRzRzR committed 265 days ago

Verified 4854dbf9

Fix loading of Qwen3 FP8 (#43494)

githubnemo committed 265 days ago

Verified a1f63d55

fix `ShieldGemma2IntegrationTest::test_model` (#43343)

sywangyi committed 265 days ago

Verified 2d5a6d03

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

sywangyi committed 265 days ago

Verified 461cfb14

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

ydshieh committed 266 days ago

Verified dfe30827

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

raimbekovm committed 266 days ago

Verified fa5dfe06

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

zucchini-nlp committed 266 days ago

Verified 3cbf8471

Older