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 supported54 Subscribers
Add a CodeTriage badge to transformers
Help out
- Issues
- Keep float32 MoE router logits in float32 under autocast
- Trainer: cast the model back in train() after an fp16/bf16 full eval
- [serge] Fix 2 integration tests for model `hrm_text` failing with `output_mismatch` (tensor values differ (2))
- Gemma3n: Flash Attention 2 vs eager attention mismatch in 3 tests (2.1% elements differ, max abs diff 0.109)
- [LUKE] Tie the masked LM decoder to the word embeddings
- [serge] Fix 2 integration tests for model `seamless_m4t` failing with `other`, raising `TypeError` (other (2))
- feat: expose opt-in assisted decoding acceptance metrics
- Add head-sharded TP for Qwen3.5 Gated DeltaNet
- [hardware support] Moore Threads GPU (MUSA) support in flash linear attention
- Fix error message and remove duplicate code in convert_slow_tokenizer.py
- Docs
- Python not yet supported