Rigtune
Open-source tool that benchmarks llama.cpp, vLLM and SGLang settings on your GPU and traffic, then serves the winner behind an OpenAI-compatible API.
- AI tool
- Free
Overview
Rigtune is an open-source tool that measures which LLM serving settings work best on your own GPU and traffic. It benchmarks configurations for llama.cpp, vLLM and SGLang against your hardware and workload. It then serves the winning setup behind one OpenAI-compatible API. It is licensed under MIT. You get the best-measured serving setup for your hardware, running and ready to call.
Key features
- Benchmarks serving settings on your hardware
- llama.cpp, vLLM and SGLang
- Tests against your own workload
- Serves the winner
- OpenAI-compatible API
- Open source under MIT
Use cases
Show all 7Show fewer
Alternatives
Last updated 2026-11-21