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
- [Usage]: vllm does not return content for vicuna
- [Bug]: Qwen-14B-Chat-Int4 with guided_json error
- [Usage]: vllm as typically workload on varies silicon to do performance analysis
- [Installation]: missing dependencies in neuron requirements
- Can you guys give the implementation for chat conversation based on current chat and chat history using vllm and llama-2-13B model
- [Bug]: When enforce_eager==False, the GPU memory is not cleaned by the standard cleanup() procedure
- [Bug]: leading space within content via OpenAI Compatible Server
- FP8 MoE implementation, take 2
- [Bug]: mixtral 8x7b outputs in-consistency between vllm and HF transfomers
- [Performance]: Can we merge the lora shrink and expand kernels in up_proj?
- Docs
- Python not yet supported