JetBrains Releases Mellum2.1 Small-Scale MoE Model

JetBrains released its Mellum2.1 model collection on the LocalLLaMA community, including a Thinking-GGUF version with 12B total parameters and 2.5B active parameters in a MoE architecture.

• JetBrains shares RAG pipeline for semantic code search2026-10-04
• JetBrains Air: A System of Products for Agentic Software Development2026-09-22
• RTK claims token savings, benchmarks disagree2026-09-11
• HN thread urges resisting 'AI'2026-09-11