Muse Glimmer is Meta's 30B open-weight model for agentic tasks

Meta also claims Glimmer delivers state-of-the-art results for its size class on tool use, multi-step planning and agentic task completion, naming Gemma4-31B and Qwen3.6-27B as comparisons. The page says Meta is releasing the model under the Apache 2.0 license. The performance comparisons and licensing statement come from Meta; neither was independently verified here.

Many accounts welcomed Meta's Muse Glimmer report for its detailed distillation methods and open 30B weights runnable on consumer GPUs, while some replies dismissed the simulated digital lives as unrealistic or overly fictional.

Based on 156 sentiment-bearing replies from 145 accounts across 8 conversations.

Many accounts welcomed Meta's Muse Glimmer report for its detailed distillation methods and open 30B weights runnable on consumer GPUs, while some replies dismissed the simulated digital lives as unrealistic or overly fictional.

Based on 156 sentiment-bearing replies from 145 accounts across 8 conversations.