huggingface/transformers

Pull Requests Commits

Merge branch 'main' into fix_moe_for_vllm

ArthurZucker committed 12 days ago

Verified b46af7f7

Delete irrelevant test that sometimes fails (#42515)

Rocketknight1 committed 12 days ago

Verified 305f82a1

[loading] `tie_weights` is finally symmetric (#42507)

Cyrilvallez committed 12 days ago

Verified 9f587100

perf: Remove implicit CPU-GPU syncs due to implicit .item() call (#42433)

Dhruv88 committed 12 days ago

Verified a3881a8d

Fix broken link to dataset food101 (#42495)

JuanFKurucz committed 12 days ago

Verified 297f5ae8

Small migration guide update (#42510)

ArthurZucker committed 12 days ago

Verified f78723fa

Fix a bug where compile_config gets deleted (#42474)

remi-or committed 12 days ago

Verified a0216a80

Patch Mistral Common Tokenizer 2 (#41962)

juliendenize committed 12 days ago

Verified aaa73254

Handle decorator with optional arguments better (#42512)

hmellor committed 12 days ago

Verified 2dba9725

logic to select tf32 API as per Pytorch version (#42428)

khushali9 committed 12 days ago

Verified 7f5c2094

[`CI`] Fix copies (#42487)

vasqu committed 12 days ago

Verified 01823d7e

remove tokenizer warning (#42483)

SunMarc committed 12 days ago

Verified e51e75e1

Fix `pr_build_doc_with_comment.yml` not reporting correctly (#42505)

ydshieh committed 12 days ago

Verified 0c05c3bd

Convenient default behavior for pipeline TTS usage. (#42473)

ebezzam committed 12 days ago

Verified 26dbe643

Add afmoe model (#42168)

alyosha-swamy committed 14 days ago

Verified cac0a28c

Fix broken models due to modeling/weight mapping (#42468)

Cyrilvallez committed 15 days ago

Verified 2a61590a

[`Docs`] Fix build on other langs (#42485)

vasqu committed 15 days ago

Verified a4a7008e

add rotary kernel support to Qwen3 model (#41147)

kaixuanliu committed 15 days ago

Verified 6587d77f

Fix minor bugs in ConversionOps and WeightConverter (#42479)

SunMarc committed 15 days ago

Verified 4abfd165

Fix gpt2 tokenizer `add_prefix_space` default value (#42481)

SunMarc committed 15 days ago

Verified c79ac071

[quantization] make torchao tests slow (#42482)

MekkCyber committed 15 days ago

Verified f8e29b5a

Skip failing irrelevant test for ColQwen2 (#42480)

Rocketknight1 committed 15 days ago

Verified 7b5f1db8

add require_deterministic_for_xpu to make the case pass in xpu (#42439)

sywangyi committed 15 days ago

Verified 98fc2a22

extend CwmIntegrationTest to xpu (#42314)

sywangyi committed 15 days ago

Verified 5dcda1b9

Fix doc builds (#42478)

Rocketknight1 committed 15 days ago

Verified 230c3234

🚨 Move `rotary_partial_emb` to RopeParams and delete unnecessary code 🔪 (#42255)

zucchini-nlp committed 15 days ago

Verified 078ff685

Update migration guide - generation config (#42470)

zucchini-nlp committed 15 days ago

Verified 554fb40e

replace source_keys and target_keys (#42471)

SunMarc committed 15 days ago

Verified b54f7907

[Trainer] use output.loss when using liger-kernel (#42444)

kashif committed 15 days ago

Verified 6db43321

remove reference to TF models from docs (#42443)

zucchini-nlp committed 15 days ago

Verified c95d4aff

Older