Muse Glimmer 30B

Release Summary

Meta's 30B open-weight multimodal agent model distilled from Muse Spark for local coding, tool use, and always-on device agents. Apache 2.0 weights, 131K context, text and image input, and a DFlash drafter for speculative decoding on a single consumer GPU. Meta reports 51.2 on SWE-bench Pro, 76.0 on SWE-bench Verified, and 51.7 on Terminal-Bench 2.1 at high reasoning versus Gemma 4 31B and Qwen3.6-27B.

Timeline