vllm
https://github.com/vllm-project/vllm
Python
A high-throughput and memory-efficient inference and serving engine for 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 supported3 Subscribers
Add a CodeTriage badge to vllm
Help out
- Issues
- Sequoia: Scalable, Robust, and Hardware-aware Speculative Decoding
- [Bug]: Incompatible version between torch and triton
- [Frontend] support new lora module to a live server in OpenAI Entrypoints
- [Bugfix] Fix beam search logits processor
- [Feature]: DeepSpeed-FP6: An Optimization Approach from Microsoft,6-bit Floating Point (FP6)
- [Usage]: Trying to run VLLM on L4 machine, but got "Expected all tensors to be on the same device" exception
- Dynamic Multi LoRA Load \ Delete Support
- [Feature]: vLLM does not work with Hugging Face ZeroGPU Spaces
- [Usage]: No tokenizer while ray usage
- [Feature]: Support distributing serving with KubeRay's autoscaler
- Docs
- Python not yet supported