JetBrains Releases Mellum2.1: A 12B MoE Open Model for Coding Agents
What happened
JetBrains released Mellum2.1, an Apache 2.0, 12B mixture-of-experts thinking model with 2.5B active parameters. JetBrains has released Mellum2.1 , an open model built for coding agents and fast sub-agents.
It ships under Apache 2.0 on Hugging Face . The upgrade comes almost entirely from reinforcement learning (RL) in real software environments.
RL in real repositories lifted its SWE-bench Verified score from 2.0 to 47.0. Mellum2.1 is a 12B mixture-of-experts thinking model from JetBrains that activates 2.5B parameters per token.
Key facts
- JetBrains โ released: Mellum2.1, an Apache 2.0, 12B mixture-of-experts thinking model with 2.5B active parameters
- JetBrains has โ released: Mellum2.1 , an open model built for coding agents and fast sub-agents
Sources & evidence
- MarkTechPost Reporting source
JetBrains Releases Mellum2.1: A 12B MoE Open Model for Coding Agents โ
https://www.marktechpost.com/2026/10/08/jetbrains-releases-mellum2-1-a-12b-moe-open-model-for-coding-agents/