Skip to content

Issues: huggingface/transformers

[Community contributions] Model cards
#36979 opened Mar 25, 2025 by stevhliu
Open 69
Beta
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

1,057 Open 16,069 Closed
1,057 Open 16,069 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
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Issues list

AutoModel cant load Qwen/Qwen2.5-0mni-7B bug
#37794 opened Apr 25, 2025 by liwenju0
4 tasks
Llama4 config is not compatible with model initialization bug
#37793 opened Apr 25, 2025 by fingertap
3 of 4 tasks
1
Whisper chunking algorithm increases WER bug
#37789 opened Apr 25, 2025 by asusdisciple
4 tasks
Loading a Pytorch model from a Tensorflow saved model doesn't work bug
#37786 opened Apr 25, 2025 by arjunaskykok
2 of 4 tasks
Support multimodal models in vLLM with transformers backend External Using the library with external tools (onnx, tflite, ...) VLM
#37780 opened Apr 25, 2025 by zucchini-nlp
2 of 7 tasks
@zucchini-nlp
[Trainer] tot update steps is incorrect bug
#37777 opened Apr 25, 2025 by efsotr
2 of 4 tasks
1
1
error: subprocess-exited-with-error when install transformerspython bug
#37775 opened Apr 25, 2025 by vfdff
4 tasks
1
Force the model to write some tokens mid-generation? Feature request Request for a new feature
#37771 opened Apr 24, 2025 by blazgocompany
2
[Trainer] As gradient_accumulation_steps increases, the loss also increases bug
#37766 opened Apr 24, 2025 by efsotr
2 of 4 tasks
2
Performance degradation on certain vision models from v4.51.* bug
#37748 opened Apr 24, 2025 by yuan-thomas
3 of 4 tasks
1
Add MM Grounding DINO New model
#37744 opened Apr 24, 2025 by rziga
2 tasks done
1
Failed to load model with transformers 4.51.3 when WORLD_SIZE set to 1 on nvidia gpu bug
#37737 opened Apr 24, 2025 by nv-guomingz
4 tasks
8
ValueError: GGUF model with architecture deci is not supported yet. Feature request Request for a new feature
#37736 opened Apr 24, 2025 by iEddie-cmd
2
tokenizer is still being used in Trainer instead of processing_class
#37734 opened Apr 24, 2025 by arjunaskykok
1
3
Facing "NotImplementedError: config.do_stable_layer_norm is False is currently not supported" in FlaxWav2vec2Models bug Flax
#37731 opened Apr 24, 2025 by premmurugan229
2 of 4 tasks
1
Gemma3 is ExecuTorch compatible Feature request Request for a new feature
#37727 opened Apr 23, 2025 by guangy10
Make argmax in post_process_semantic_segmentation optional Feature request Request for a new feature
#37715 opened Apr 23, 2025 by simonreise
1
Loading and Saving Pretrained model to the same directory raises SafeTensorError: IOError bug
#37713 opened Apr 23, 2025 by minerharry
3 of 4 tasks
3
Very slow model instantiation bug
#37712 opened Apr 23, 2025 by rvorias
4 tasks
@Rocketknight1
7
Can't perform inference with images on Gemma-3-12b-it-qat-int4.0 bug
#37710 opened Apr 23, 2025 by njemanzedavid
2 of 4 tasks
1
last_cache_position definition issue in hybrid SWA models bug
#37706 opened Apr 23, 2025 by plienhar
1
[i18n-Chinese] Translating model_doc/bert.md to Chinese WIP Label your PR/Issue with WIP for some long outstanding Issues/PRs that are work in progress
#37705 opened Apr 23, 2025 by Nanji-Huaji
2
Tokenizing with apply_chat_template behaves differently from regular tokenizing bug
#37686 opened Apr 22, 2025 by sayanshaw24
2 of 4 tasks
1
Behaviour of batch_eval_metrics determines the include_for_metrics behaviour bug
#37683 opened Apr 22, 2025 by prabhuteja12
1 of 4 tasks
4
Model.from_pretrained breaks when using SinusoidalEmbedding bug
#37671 opened Apr 22, 2025 by ZhiyuanChen
4 tasks
ProTip! no:milestone will show everything without a milestone.