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
- Allow max_batch_prefill_tokens to be less than max_input_length
- Getting error with LoRA adapters for Qwen 2.5 0.5B Instruct
- Quantization appears to be broken, at least for AWQ and BnB
- Not able to preload local adapters
- int: Enable manual trigger for server tests wf
- Attention not working properly in FlashRobertaModel and FlashBertModel
- why prune in mergeStrategy no rescale remaining?
- Not able to host Llama3.2-11b on Azure A100 80GB server
- Cannot run a FP8 quantized model with LoraX
- Throughput and Latency degradation with a single LoRA adapter on A100 40 GB
- Docs
- Python not yet supported