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 supported13 Subscribers
Add a CodeTriage badge to vllm
Help out
- Issues
- [CI Failure]: torch.compile cache are reused across unittests.
- [P/D][Metrics] Consider combined/summed metrics (e.g. ttft and e2e_request_latency) for prefill and decode instances
- [Bug]: Olmo-3 vLLM Online API generates gibberish
- [RFC]: Detailed Usage Guide for AFD Support in vLLM
- [Performance]: Add optimized fused-moe kernels for low bit quantized gguf models
- fix hf vllm mapping for mixtral that was breaking due to v5 coming
- [Bug]: Duplicate registration of a fake implementation for the gptq_marlin_repack operator causing vllm serve to fail.
- [Bug]: FP8 KV Cache fails for google/gemma-3-1b-it on Hopper with backend FlashInfer
- [Bug]: "expandable_segments: True" causes vLLM EngineCore initialization to fail when running Qwen3 VL models
- [Bug]: "\n\n" content between reasoning and tool_call content when tool_call and stream mode
- Docs
- Python not yet supported