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
- [Feature]: Support attention backend with FlexAttention
- [Usage]: 流式输出前面几个字符为啥要设置成空字符?不能直接输出模型的生成吗
- Enable FusedSDPA for prompt attention with env VLLM_PREFILL_USE_FUSEDSDPA=true
- [Model] activation gpu mem reuse in model forward
- [FrontEnd] Keep RPC server tcp protocol
- [Feature]: Improve the compile times of gptq_marlin.cu
- [Bug]: Extra body don't work when response_format is also sent for serving.
- [Feature]: Beam Search also requires diversity
- [Bug]: vllm is hang after upgrade to v0.5.4
- [Usage]: Acceptance rate for Speculative Decoding
- Docs
- Python not yet supported