Skip to content

Pull requests: huggingface/transformers

New pull request New
614 Open 18,766 Closed
614 Open 18,766 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

chore: fix typos in tests directory
#36785 opened Mar 18, 2025 by threewebcode Loading…
2 of 5 tasks
Fix attention_mask dimension issue in GPT2Model
#36782 opened Mar 18, 2025 by sukwoojang Loading…
2 of 5 tasks
1
1
Use public export API on torch 2.5 and future
#36781 opened Mar 18, 2025 by guangy10 Loading…
1 task done
1
Export for Phi4-mini
#36780 opened Mar 18, 2025 by guangy10 Loading…
3 tasks done
1
Proposed fix for V4.49.0 gemma 3 release to solve issue #36758
#36779 opened Mar 17, 2025 by Zephyr271828 Loading…
2 of 5 tasks
1
Registering New Tokenizer + Possible Future Foundational Model Architecture Added to Transformers
#36778 opened Mar 17, 2025 by yuchenxie4645 Loading…
2
Fixing typo in gemma3 image_processor_fast and adding a small test
#36776 opened Mar 17, 2025 by Zebz13 Loading…
4 of 5 tasks
1
1
rewrite main method in Qwen2, making it more clear
#36772 opened Mar 17, 2025 by ailunc • Draft
4 of 5 tasks
1
fix hqq due to recent modeling changes
#36771 opened Mar 17, 2025 by SunMarc Loading…
2
Add support for audios in apply_chat_template
#36770 opened Mar 17, 2025 by junnei Loading…
5 tasks
2
Fix import for torch 2.0, 2.1 - guard typehint for "device_mesh"
#36768 opened Mar 17, 2025 by qubvel Loading…
1
Update min version of safetensors to allow integer indexing
#36767 opened Mar 17, 2025 by SunMarc Loading…
3
🌐 [i18n-KO] Translated electra.md to Korean
#36763 opened Mar 17, 2025 by Kim-Ju-won • Draft
6 of 10 tasks
1
feat: expose the strict flag to allow catching missing model layers while loading a checkpoint
#36760 opened Mar 17, 2025 by HarikrishnanBalagopal Loading…
3 tasks
2
Fix: remove the redundant snippet of _whole_word_mask
#36759 opened Mar 17, 2025 by HuangBugWei Loading…
1 of 5 tasks
1
🌐 [i18n-KO] Translated gpu_selection.md to Korean
#36757 opened Mar 17, 2025 by nsbg • Draft
6 of 10 tasks
6
🌐 [i18n-KO] Translated 'serving.md' to Korean
#36756 opened Mar 17, 2025 by olccihyeon • Draft
10 tasks
Pass num_items_in_batch directly to loss computation
#36753 opened Mar 16, 2025 by eljandoubi Loading…
1
1
Add Qwen2.5-Omni
#36752 opened Mar 16, 2025 by BakerBunker Loading…
4 of 5 tasks
1
🌐 [i18n-KO] Translated qwen2_vl.md to Korean
#36750 opened Mar 16, 2025 by MinJu-Ha • Draft
5 of 10 tasks
13
Remove extra tensor clone in PyTorch code
#36748 opened Mar 16, 2025 by cyyever Loading…
2 of 5 tasks
1
Loading optimizations
#36742 opened Mar 15, 2025 by Cyrilvallez Loading…
3
[gemma 3] multimodal checkpoints + AutoModelForCausalLM
#36741 opened Mar 15, 2025 by gante Loading…
8
[WP] PagedAttention + Prefix Cache for FlashAttention2
#36737 opened Mar 15, 2025 by huseinzol05 • Draft
1
Convert _VALID_DICT_FIELDS to class attribute for shared dict parsing in subclasses
#36736 opened Mar 15, 2025 by Tavish9 Loading…
2 of 5 tasks
2
ProTip! no:milestone will show everything without a milestone.