Meta releases Muse Glimmer weights for local agents
Meta Superintelligence Labs has put Muse Glimmer, a 30-billion-parameter model, on Hugging Face under Apache 2.0, weights and docs included. The target is always-on agents running on a Mac or a PC with a single consumer GPU: the model plans tasks, calls tools, and recovers when those tools fail. It handles more than 100 languages and has switchable reasoning depth.
At full precision it needs more than 55 GB of memory. Quantization squeezes the language model below 20 GB. For the K-Quant-17GB build, Meta claims 3.1x faster decoding on an RTX 5090, 1.8x on an M5 Max, and 1.5x on an M4 Max.
Meta benchmarks the results against Gemma4-31B and Qwen3.6-27B. Mark Zuckerberg said the weights for Muse Spark 1.2 are coming soon.
Related stories
- 3.1% word error rate leads the streaming speech chart
- Xiaomi's new MiMo models carry an unverified top-6 claim
- Cohere claims a WMT26 lead with a non-reasoning translator
- 552B MoE opens DeepSeek's new architecture family
- Six models, 0.9B to 375B, ship with training logs
- DeepSWE 1.1 score puts Muse Spark 1.3 above Opus 5
Comments
No comments yet. Be the first.
Join the conversation
Sign in with Google to leave a comment. Your name and avatar come from your Google profile, and the comment appears after moderation.