transformers
https://github.com/huggingface/transformers
Python
🤗 Transformers: State-of-the-art Machine Learning for Pytorch, TensorFlow, and JAX.
Triage Issues!
When you volunteer to triage issues, you'll receive an email each day with a link to an open issue that needs help in this project. You'll also receive instructions on how to triage issues.
Triage Docs!
Receive a documented method or class from your favorite GitHub repos in your inbox every day. If you're really pro, receive undocumented methods or classes and supercharge your commit history.
Python not yet supported55 Subscribers
View all SubscribersAdd a CodeTriage badge to transformers
Help out
- Issues
- Fix Whisper beam_indices so compute_transition_scores does not select the beam twice
- [Feature request] FP8 KV cache + flash-attention: is this in scope?
- Trainer fix for DTensor API
- `compute_transition_scores()` applies beam selection twice for Whisper → out-of-bounds gather
- Prefetch only the byte spans each rank will read
- [WIP] [MiniCPMV4_6] Fix vit_merger crash when video frames are sliced (non-uniform target_sizes)
- Trainer: make checkpoints work under tensor parallelism
- Add image-text contrastive loss to Blip2ForImageTextRetrieval
- `Trainer` deletes `config.bos_token_id` when the tokenizer declares no BOS, and persists it
- fix(integrations): print the base model architecture for PEFT-wrapped models
- Docs
- Python not yet supported