Mellum by JetBrains
Fast language models built for modern AI workflows
Mellum is JetBrains' family of open language models built for ultra fast inference and high performance AI workloads. Designed for software engineering and production environments, it delivers low latency, efficient compute, and lower operating costs for code generation, RAG pipelines, AI agents, and real time applications. Available as open source, Mellum can run locally or in the cloud, giving developers complete flexibility, privacy, and infrastructure control.