Skip to content

Pull requests: huggingface/transformers

New pull request New
1,034 Open 22,746 Closed
1,034 Open 22,746 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
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

Remove Neptune integration references and deprecate NeptuneCallback
#42666 opened Dec 5, 2025 by qgallouedec Loading…
1
3
Some optimizations for offloading
#42665 opened Dec 5, 2025 by Cyrilvallez Loading…
1
FIX Error when trying to load non-LoRA PEFT
#42663 opened Dec 5, 2025 by BenjaminBossan Loading…
2 of 5 tasks
2
[mRope] Fix warnings
#42660 opened Dec 5, 2025 by vasqu Loading…
2
Use hfh's is_offline_mode helper
#42657 opened Dec 5, 2025 by Wauplin Loading…
4
New Feature: Enabling Speculative Decoding with Batch Size > 1 (If draft and target model share tokenizer)
#42655 opened Dec 5, 2025 by YanivDorGalron Loading…
[Quantization] Fix FP8 experts replacing
#42654 opened Dec 5, 2025 by MekkCyber Loading…
2
[Ernie 4.5 Moe] Fix routing, weights, and update expectations
#42653 opened Dec 5, 2025 by vasqu Loading…
6
Only default rope_parameters to empty dict if there is something to put in it
#42651 opened Dec 5, 2025 by hmellor Loading…
5
qwen2 tok regression
#42644 opened Dec 5, 2025 by itazap Loading…
3
Command-a-vision fix
#42642 opened Dec 5, 2025 by dongluw Loading…
5 tasks
4
Xpu fp8 gem
#42639 opened Dec 5, 2025 by sywangyi Loading…
2
[docs] WeightConverter
#42636 opened Dec 4, 2025 by stevhliu Loading…
1
allow registration of custom checkpoint conversion mappings
#42634 opened Dec 4, 2025 by winglian Loading…
5 tasks
1
Make GraniteMoeHybridModel compatible with torch.export
#42631 opened Dec 4, 2025 by juanigp Loading…
5 tasks
3
Remove redundant else in activation.py
#42621 opened Dec 4, 2025 by leaderofARS Loading…
fix: support tensor labels in DataCollatorWithFlattening
#42620 opened Dec 4, 2025 by hqkqn32 Loading…
4 tasks done
1
3
update and add Expectations for mistral3/internvl tests
#42616 opened Dec 4, 2025 by Abdennacer-Badaoui Loading…
7
[core] fix fp-quant
#42613 opened Dec 4, 2025 by MekkCyber Loading…
4
Let transformers know when a model is being traced via jax.jit (torchax)
#42611 opened Dec 3, 2025 by qihqi Loading…
4 tasks
1
5
Raise error when missing or incorrect dates
#42610 opened Dec 3, 2025 by yonigozlan Loading…
2
Fix dimensional mismatch in Mistral3MultiModalProjector for multiple vision feature layers
#42602 opened Dec 3, 2025 by arrdel Loading…
1
5
[pipeline] fix unwanted import failure
#42598 opened Dec 3, 2025 by eustlb Loading…
1
1
Fix convert_tekken_tokenizer
#42592 opened Dec 3, 2025 by juliendenize Loading…
5 tasks
3
Document the /v1/models endpoint
#42588 opened Dec 3, 2025 by LysandreJik Loading…
7
ProTip! no:milestone will show everything without a milestone.