Back to AIBriefs
LaunchAI ModelsDevelopers

JetBrains open-sources Mellum2, a 12B MoE model

JetBrains open-sources Mellum2, a 12B MoE modelblog.jetbrains.com

Mellum2 is a 12B-parameter Mixture-of-Experts model with only 2.5B active parameters per token, released under Apache 2.0. It is specialized for code and natural language, achieving competitive benchmark results with less than half the inference time of similar-sized models.

RT @jetbrains: Mellum started with code completion. Mellum2 is built for more – handling both...16 days agoHugging Face
··Discuss
16 days ago
JetBrains open-sources Mellum2, a 12B MoE model — AIBriefs