telos

Open-source token optimization engine for AI workloads. Minimize LLM inference costs while preserving output quality