huggingface/transformers

Pull Requests Commits

[CI] AMD docker test: restore DockerHub login

Abdennacer-Badaoui committed 4 days ago

1341c925

[CI] AMD docker: bump base to rocm7.2.2/pytorch2.10, use pre-built FA wheel

Abdennacer-Badaoui committed 4 days ago

d8551b8d

fix(minicpmv4_6): skip invalid failing tests (#45836)

kaixuanliu committed 4 days ago

Verified b75feb2a

audio tester class (#45391)

tarekziade committed 4 days ago

Verified 83f33cda

Remove deprecation cycle for inputs embeds (#45885)

Cyrilvallez committed 4 days ago

Verified 2d6815e4

Warn about forgetting attention mask functions (#45811)

TuringTux committed 4 days ago

Verified b890df15

Remove deprecation cycle for `cache_position` in masking primitives (#45884)

Cyrilvallez committed 4 days ago

Verified 6c66de3f

Revert "[deepseek_v4] fix CSA per-query masking in eager path"

ArthurZucker committed 4 days ago

436ba195

Revert "Merge branch 'main' of github.com:huggingface/transformers into fix-deepseek-v4-csa-per-query-mask"

ArthurZucker committed 4 days ago

d996062a

Merge branch 'main' of github.com:huggingface/transformers into fix-deepseek-v4-csa-per-query-mask

ArthurZucker committed 4 days ago

e9fd66d1

[deepseek_v4] fix CSA per-query masking in eager path

ArthurZucker committed 4 days ago

ac372e10

fix(testing_utils): guard get_device_capability with torch.cuda.is_available() (#45351)

RudrenduPaul committed 4 days ago

Verified 1ee03164

🚨 Refactor ViT to updated standards (#41693)

yonigozlan committed 6 days ago

Verified c9de1097

fix(rf_detr): fix failing tests (#45845)

kaixuanliu committed 7 days ago

Verified 381032b7

fix(granite4_vision): auto-fix failing tests (#45844)

kaixuanliu committed 7 days ago

Verified 4f076084

fix(laguna): fix failing tests (#45842)

kaixuanliu committed 7 days ago

Verified 9f0b8aee

:rotating_light: Generic Sequence Classifier works for multimodal models (#44664)

zucchini-nlp committed 7 days ago

Verified 1d5296f2

Add Qwen3.5 support for token classification (#45833)

Ratwood98 committed 7 days ago

Verified 0314464f

Fix import error in moe.py by providing explicit schema to custom_op (#45804)

rraghavkaushik committed 7 days ago

Verified a5cc128d

fix: correct typo 'seperate' -> 'separate' in comments across multipl… (#45824)

Caromisc committed 7 days ago

Verified 4b23cfc9

🚨 [Fuyu] Remove FuyuBatchFeature subclass, use BatchFeature with skip_tensor_conversion (#45808)

Abineshabee committed 7 days ago

Verified 44a2e7a4

Keep deleting (#45802)

zucchini-nlp committed 7 days ago

Verified 9ea2afbc

Fix gemma4 with multi-gpu setup (#45826)

zucchini-nlp committed 7 days ago

Verified 95933eb6

Add HyperCLOVAX SEED Think 14B (#44956)

bigshanedogg committed 8 days ago

Verified 0f90c8e3

Fix `kernelize()` crash for gpt_oss: missing `@use_kernel_func_from_hub` on `apply_rotary_pos_emb` (#45799)

jiqing-feng committed 8 days ago

Verified ebac5e52

Cache `merged_typed_dict` to not break `validate_typed_dict` caching (#45807)

lgeiger committed 8 days ago

Verified 95fdeb31

Get rid of deprecated use_return_dict call. (#45815)

rasmi committed 8 days ago

Verified 28ab024e

fix(qianfan_ocr): add XPU expectations (#45615)

kaixuanliu committed 8 days ago

Verified 15cef23c

Fix shared config mutation issue in flash_attn_from_config (#45678)

kaixuanliu committed 8 days ago

Verified 10f53971

Add RF-DETR (#36895)

sbucaille committed 8 days ago

Verified 81fe4f16

Older