lorax
https://github.com/predibase/lorax
Python
Multi-LoRA inference server that scales to 1000s of fine-tuned LLMs
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 supported1 Subscribers
Add a CodeTriage badge to lorax
Help out
- Issues
- (WIP) Support targeting the embedding layer for LoRA
- Always consider base model (no adapter) to be active
- Panic when adapter cannot be loaded
- Add RoPE scaling CLI args
- Extend testing
- Support multiple ranks per SGMV op
- Skip the download process from `lorax-launcher` if model weights already on local disk
- Add per-adapter metrics to `/metrics` endpoint
- Added list of exported metrics
- sync.sh script fails for some models (Llama-2-70b being one of them)
- Docs
- Python not yet supported