huggingface/transformers

Pull Requests Commits

Merge branch 'main' into fix_deepseek_v2_experts

Rocketknight1 committed 9 hours ago

Verified f92c5535

Fix eager experts for DeepSeekV2

Rocketknight1 committed 9 hours ago

5f75d26d

[CB] Ensure parallel decoding test passes using FA (#43277)

remi-or committed 10 hours ago

Verified 61317f5a

Fix failing `PegasusX`, `Mvp` & `LED` model integration tests (#43245)

Sai-Suraj-27 committed 11 hours ago

Verified 1efe1a63

[consistency] Ensure models are added to the `_toctree.yml` (#43264)

Cyrilvallez committed 15 hours ago

Verified e8ae3731

[docs] tensorrt-llm (#43176)

stevhliu committed 1 day ago

Verified c85be989

[style] Fix init isort and align makefile and CI (#43260)

Cyrilvallez committed 1 day ago

Verified 38022fd8

Fix failing `Hiera`, `SwiftFormer` & `LED` Model integration tests (#43225)

Sai-Suraj-27 committed 1 day ago

Verified e977446e

fix `_get_test_info` in `testing_utils.py` (#43259)

ydshieh committed 1 day ago

Verified d79b0b88

Remove deprecated and unused `position_ids` in all `apply_rotary_pos_emb` (#43255)

Cyrilvallez committed 1 day ago

Verified 159219a6

Fix failing `Vip-llava` model integration test (#43252)

Sai-Suraj-27 committed 1 day ago

Verified ad7f4d01

Adding to run slow (#43250)

IlyasMoutawwakil committed 1 day ago

Verified 35c26ace

Fix benchmark script (#43253)

Cyrilvallez committed 1 day ago

Verified b40b7a76

Make benchmark-v2 to be device agnostic, to support more torch built-in devices like xpu (#43153)

yao-matrix committed 1 day ago

Verified 3ef2e3ba

[CB] Soft-reset offloading (#43150)

remi-or committed 1 day ago

Verified b6a473d0

Remove more deprecated objects/args (#43195)

Cyrilvallez committed 1 day ago

Verified 866bf24b

[GLM-Image] AR Model Support for GLM-Image (#43100)

zRzRzRzRzRzRzR committed 1 day ago

Verified 136c621c

[GLM-4.7] GLM-Lite Supoort (#43031)

zRzRzRzRzRzRzR committed 1 day ago

Verified 76732b4e

Deprecate dtype per sub config (#42990)

zucchini-nlp committed 1 day ago

Verified 7918b152

Qwen-VL video processor accepts min/max pixels (#43228)

zucchini-nlp committed 1 day ago

Verified acb0f226

Fix typo: seperately → separately in lw_detr converter (#43235)

skyvanguard committed 1 day ago

Verified c0d2e26f

docs: fix grammatical error in README.md (#43236)

davidfertube committed 1 day ago

Verified 15f89963

More robust diff checks in `add_dates` (#43199)

yonigozlan committed 2 days ago

Verified 3aa89c07

Reapply modular to examples (#43234)

Cyrilvallez committed 2 days ago

Verified 7f20ad00

Fix unsafe torch.load() in _load_rng_state allowing arbitrary code execution (#43140)

ColeMurray committed 2 days ago

Verified 03c8082b

Remove unused imports (#43078)

cyyever committed 2 days ago

Verified 2aa7b652

Add LWDetr model (#40991)

sbucaille committed 2 days ago

Verified d8541c36

Fix flashattn wrt quantized models (#43145)

SunMarc committed 2 days ago

Verified 35fe3419

Fix failing `Owlv2ModelIntegrationTest` & `OwlViTModelIntegrationTest` (#43182)

Sai-Suraj-27 committed 2 days ago

Verified 61d7f8a4

Use `HF_TOKEN` directly and remove `require_read_token` (#43233)

ydshieh committed 2 days ago

Verified 5f70e191

Older