Pull Requests huggingface/transformers

fix(models): Unpack BitNet packed weights in quantizer post-loading hook to fix CI failure

#43721 opened 2026-02-03 20:01 by harshaljanjani

Add id and resume support to SwanLab integration

#43719 opened 2026-02-03 18:59 by i-pj

Fix resume_from_checkpoint ignoring user's batch size configuration

#43718 opened 2026-02-03 18:37 by lordaarush

Add AgentMarket - Real Energy Data for AI Agents

#43715 opened 2026-02-03 17:31 by stromfee

[Trainer] Move NEFTune impl to standalone functions

#43714 opened 2026-02-03 16:35 by SunMarc

More flaky generate tests

#43713 opened 2026-02-03 15:51 by Rocketknight1

Improve typing/explanations for general model properties

#43712 opened 2026-02-03 15:41 by Cyrilvallez

Fix scheduler initialization order

#43711 opened 2026-02-03 15:14 by SunMarc

[Docs] Add docs for GLM-OCR and fix EomT-DINOv3

#43710 opened 2026-02-03 15:07 by NielsRogge

fix: `test_backend_specification` failing on `VersionComparison`

#43709 opened 2026-02-03 14:26 by tarekziade

[Model] Add SLANeXt Model Support

#43707 opened 2026-02-03 13:33 by liu-jiaxuan

Allow bi-directional attention for all models

#43705 opened 2026-02-03 11:45 by Cyrilvallez

Add moonshine

#43702 opened 2026-02-03 09:32 by eustlb

avoid using specified backend for tp tests

#43699 opened 2026-02-03 07:33 by YangKai0616

Fix T5 v1.1 detection

#43681 opened 2026-02-02 13:12 by githubnemo

🚨 Generation cache preparation

#43679 opened 2026-02-02 11:29 by zucchini-nlp

one possible fix issue #43676

#43677 opened 2026-02-02 10:31 by tarekziade

Fix pe uts

#43672 opened 2026-02-02 04:10 by YangKai0616

Add _get_signature_columns method to allow custom trainers to override column filtering

#43663 opened 2026-01-31 21:34 by abigailtech

Fix incorrect timestamp calculation in Qwen3VL Processor

#43659 opened 2026-01-31 18:57 by jonathan-fulton

Fix TypeAdapter NameError in transformers CLI

#43656 opened 2026-01-31 18:54 by jonathan-fulton

fix(tokenizer): Avert special token property overwrites in batch add_tokens calls

#43654 opened 2026-01-31 18:31 by harshaljanjani

Add _loss_is_scaled_for_ga to allow custom trainers to control gradient accumulation loss scaling

#43651 opened 2026-01-31 17:51 by abigailtech

Check new failures reporting 5

#43649 opened 2026-01-31 14:06 by ydshieh

check

#43647 opened 2026-01-31 11:38 by ydshieh

Docs: fix SigLIP2 usage examples

#43641 opened 2026-01-31 07:00 by KOKOSde

Add _metrics dict to Trainer for custom metric logging

#43636 opened 2026-01-30 21:53 by abigailtech

🚨 T5Gemma2 model structure

#43633 opened 2026-01-30 15:50 by zucchini-nlp

Add VibeVoice ASR New model Audio

#43625 opened 2026-01-30 14:06 by ebezzam

Add Promptable Visual Segmentation pipeline

#43613 opened 2026-01-30 03:00 by yonigozlan