huggingface/transformers

Pull Requests Commits

🔴 VLM: compile compatibility (#35724)

zucchini-nlp committed 359 days ago

Verified 0c78ef6c

Guard against unset resolved_archive_file (#35628)

dmlap committed 359 days ago

Verified b45cf0e9

Revert qwen2 breaking changes related to attention refactor (#36162)

ArthurZucker committed 359 days ago

Verified 96f01a36

Add require_read_token to fp8 tests (#36189)

MekkCyber committed 359 days ago

Verified cb586a39

New HIGGS quantization interfaces, JIT kernel compilation support. (#36148)

BlackSamorez committed 359 days ago

Verified 5f726f8b

Prepare processors for VideoLLMs (#36149)

zucchini-nlp committed 359 days ago

Verified 15ec971b

Add ImageProcessorFast to Qwen2.5-VL processor (#36164)

Isotr0py committed 359 days ago

Verified 33d1d715

Chat template docs (#36163)

zucchini-nlp committed 359 days ago

Verified 1931a351

CI: fix `test-save-trainer` (#36191)

zucchini-nlp committed 359 days ago

Verified 3bf02cf4

Add support for partial rotary embeddings in Phi3 model (#35947)

garg-amit committed 359 days ago

Verified 0ae93d31

Uniformize OwlViT and Owlv2 processors (#35700)

yonigozlan committed 359 days ago

Verified 336dc69d

Fix make_batched_videos and add tests (#36143)

yonigozlan committed 359 days ago

Verified e6a79817

Fix a mistake in #36175 (#36179)

ydshieh committed 359 days ago

Verified 8fd4bc7d

Follow up to SpQR integration (#36176)

MekkCyber committed 360 days ago

Verified b1a2de07

Fix the key name for _load_rng_state under torch.cuda (#36138)

wizyoung committed 360 days ago

Verified 12962fe8

Make `check_repository_consistency` run faster by MP (#36175)

ydshieh committed 360 days ago

Verified bfe46c98

Optimize Qwen2VL vision model by precomputing cos/sin embeds before ViT blocks (#35837)

li-plus committed 360 days ago

Verified 5f0fd118

Use tqdm auto (#35726)

SmartManoj committed 360 days ago

Verified d72642bc

CI: avoid human error, automatically infer generative models (#33212)

gante committed 360 days ago

Verified 62c7ea02

add disable compile option (#36161)

ArthurZucker committed 360 days ago

Verified 06231fdf

fix training issues (#36158)

ArthurZucker committed 360 days ago

Verified 0ca72592

Efficient Inference Kernel for SpQR (#34976)

elvircrn committed 360 days ago

Verified 845b0a26

Bump transformers from 4.38.0 to 4.48.0 in /examples/research_projects/adversarial (#36168)

dependabot[bot] committed 360 days ago

Verified c5506f4f

Bump transformers from 4.38.0 to 4.48.0 in /examples/tensorflow/language-modeling-tpu (#36167)

dependabot[bot] committed 360 days ago

Verified d7c5d1b5

[generate] revert change in Aria: the maximum cache length must match `max_length` (#36120)

gante committed 360 days ago

Verified 636ee574

Fix : fix doc fp8 (#36173)

MekkCyber committed 360 days ago

Verified b41591d8

Fix red CI (#36174)

ArthurZucker committed 360 days ago

Verified b079dd1f

[Modular] skip modular checks based on diff (#36130)

gante committed 360 days ago

Verified d114a6f7

Remove loading custom kernel for RT-DETRv2 (#36098)

qubvel committed 360 days ago

Verified 6397916d

Adding FP8 Quantization to transformers (#36026)

MekkCyber committed 360 days ago

Verified efe72fe2

Newer Older