Kodosumi

Deploy and scale your AI agents with Kodosumi, the open-source runtime built for developers. Fast, scalable, and fully free to use.

Kodosumi is an open-source runtime environment designed to deploy and scale AI agents efficiently. Built on Ray, Litestar, and FastAPI, it simplifies the management of long-running, complex workflows without requiring deep Ray expertise. The platform offers horizontal scalability to handle unpredictable traffic spikes and provides built-in observability with real-time monitoring via the Ray dashboard. Developers can integrate their preferred LLMs, frameworks, and tools without vendor lock-in, using a minimal YAML configuration. Kodosumi supports diverse deployment options, including Kubernetes and Docker, and is completely free to use. Ideal for those building agentic services, it balances flexibility with enterprise-grade reliability.