JetBrains Mellum2, 2배 빠른 추론 속도 달성
JetBrains가 Mellum2를 공개했습니다. 토큰당 2.5B 파라미터만 활성화하는 12B 오픈 Mixture-of-Experts 모델입니다. 이를 통해 대규모 AI 스택 내에서 저지연 라우팅, RAG, 에이전트 하위 작업이 가능해집니다.
jetbrains released mellum2, a 12b-parameter mixture-of-experts model trained from scratch on natural language and code, available under the apache 2.0 license. by activating only 2.5b parameters per token, the architecture achieves over 2x faster inference compared to similarly sized models, targeting high-frequency, low-latency tasks like prompt routing, context compression, and tool selection. the release signals a shift toward 'focal' models that handle specific agent subtasks efficiently rather than relying solely on large frontier models for every step in a pipeline. developers can now deploy this model for private coding assistance or as a lightweight orchestration layer to reduce serving costs in complex ai systems.