-
Notifications
You must be signed in to change notification settings - Fork 30.8k
Pull requests: huggingface/transformers
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
Add conditional checks to _check_and_adjust_attn_implementation()
#41542
opened Oct 13, 2025 by
zheliuyu
Loading…
[Qwen3VL] fix device mismatch error for FSDP2 training
#41536
opened Oct 12, 2025 by
HollowMan6
Loading…
1 of 5 tasks
🌐 [i18n-KO] Translated
video_processor.md
to Korean
#41531
opened Oct 12, 2025 by
chelsseeey
Loading…
5 of 10 tasks
🌐 [i18n-KO] Translated selecting.md to Korean
#41527
opened Oct 12, 2025 by
maximizemaxwell
•
Draft
5 of 10 tasks
Add max_eval_batches argument to TrainingArguments
#41524
opened Oct 11, 2025 by
KaparthyReddy
Loading…
Add test coverage for ConvNextImageProcessorFast
#41523
opened Oct 11, 2025 by
KaparthyReddy
Loading…
Fix _init_weights to safely skip int8 quantized weights
#41522
opened Oct 11, 2025 by
KaparthyReddy
Loading…
Fix forced_bos_token_id not set in generation_config
#41521
opened Oct 11, 2025 by
Addyk-24
Loading…
2 of 5 tasks
[ci] Disable workflows with secrets and custom runners to run on fork
#41515
opened Oct 10, 2025 by
HollowMan6
Loading…
1 of 5 tasks
[don't merge yet] Remove some custom datasets defined in codebase
#41511
opened Oct 10, 2025 by
ydshieh
Loading…
🌐 [i18n-KO] Translated
ko-LFM2.md
to Korean
#41502
opened Oct 10, 2025 by
ssum21
Loading…
10 tasks done
Add skip_unnecessary_grad_clip to TrainingArguments for optimized gradient clipping
#41491
opened Oct 9, 2025 by
vaibhavgarg230
Loading…
3 tasks done
Fix _init_weights to safely skip int8 tensors in Qwen2_5_VL model
#41490
opened Oct 9, 2025 by
KaparthyReddy
Loading…
🚨 [v5] Toggle the serialization format in processors
#41474
opened Oct 9, 2025 by
zucchini-nlp
Loading…
Previous Next
ProTip!
Mix and match filters to narrow down what you’re looking for.