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
- Continuous batching does not support hybrid linear-attention models (all Qwen3.5/3.6/3.8)
- Native TP breaks on Qwen3.5/3.6 hybrids: reshape error in linear-attention layers
- Native TP has no KV-head replication: tp_size > num_key_value_heads fails mid-forward
- Use the shared attention interface in encoder-decoder attention blocks
- Native TP does not shard qwen3_vl_moe experts (weights fully replicated per rank)
- Fix Moshi output fields
- [`Kernels`] Add swiglu and geglu MLP across models
- draft modular v2 + transformers format
- Handle empty FSDP shards in sharded loading: crash and hang fix
- Loading very large models across nodes is 10-30x slower than the filesystem allows (and can kill the job)"
- Docs
- Python not yet supported