Skip to content

Pull requests: huggingface/transformers

New pull request New
555 Open 18,416 Closed
555 Open 18,416 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

fix: typos in documentation files
#36122 opened Feb 10, 2025 by maximevtush Loading…
Fix potential regex catastrophic backtracking in NougatTokenizerFast
#36121 opened Feb 10, 2025 by Rocketknight1 • Draft
2
[generate] revert change in Aria: the maximum cache length must match max_length
#36120 opened Feb 10, 2025 by gante Loading…
1
update awesome-transformers.md.
#36115 opened Feb 10, 2025 by zhanluxianshen Loading…
2 of 5 tasks
3
add DeepSpeed tensor parallel initialization.
#36114 opened Feb 10, 2025 by inkcherry Loading…
Proper performant flex attention implementation
#36103 opened Feb 8, 2025 by bursteratom Loading…
3 of 15 tasks
24
Remove loading custom kernel for RT-DETRv2 cleanup Vision
#36098 opened Feb 7, 2025 by qubvel Loading…
4
Fixup another model + encoder decoder
#36095 opened Feb 7, 2025 by muellerzr • Draft
5 tasks
1
1
Add pipeline parallel plan to PretrainedConfig and PreTrainedModel
#36091 opened Feb 7, 2025 by hmellor Loading…
5 tasks
14
Fix: Llama - adjust the rotary embedding dimensions.
#36090 opened Feb 7, 2025 by abdullahselek Loading…
1 of 5 tasks
fix: dtype might change during resize
#36089 opened Feb 7, 2025 by LarsHaalck Loading…
2 of 5 tasks
5
qwen2.5vl: fix bugs when using flash2+bf16 or num_return_sequences>1
#36083 opened Feb 7, 2025 by gewenbin0992 Loading…
5 tasks
9
[docs] fix bug in deepspeed config
#36081 opened Feb 7, 2025 by faaany Loading…
2
[docs] update awq doc
#36079 opened Feb 7, 2025 by faaany Loading…
9
PyTorch's int8 & int4 WoQ GEMMs expect contiguous activations
#36078 opened Feb 7, 2025 by sanchitintel Loading…
4
[Bugfix] Fix reloading of pixtral/llava configs
#36077 opened Feb 7, 2025 by kylesayrs Loading…
7
Speedup modular conversion w/ multiproc
#36073 opened Feb 6, 2025 by muellerzr • Draft
1 of 5 tasks
4
add special tokens should probably be serialized and deserialized?
#36070 opened Feb 6, 2025 by ArthurZucker • Draft
1
1
Optim: APOLLO optimizer integration
#36062 opened Feb 6, 2025 by zhuhanqing Loading…
26
🚧 [WiP] Add Janus model
#36053 opened Feb 5, 2025 by yaswanth19 • Draft
5 tasks
1
8
Remove type hint Unpack[FlashAttentionKwargs]
#36049 opened Feb 5, 2025 by ydshieh Loading…
9
Anole add model
#36047 opened Feb 5, 2025 by zucchini-nlp Loading…
2
Fix bug in prepare_inputs_for_generation function: in Qwen2-VL (#36037)
#36038 opened Feb 4, 2025 by JamesHujy Loading…
1
Add Phi-3.5-vision
#36036 opened Feb 4, 2025 by Dahlbomii • Draft
1
[core] Large/full refactor of from_pretrained
#36033 opened Feb 4, 2025 by Cyrilvallez Loading…
1
ProTip! Type g i on any issue or pull request to go back to the issue listing page.