Meta releases Muse Glimmer, 30B open-weight model for local AI agents

Muse Glimmer ships under Apache 2.0 with a 120K+ context window and runs on a single consumer GPU (24GB VRAM) or Mac/PC. NVIDIA reports 20K tokens/sec on one GPU, and Hugging Face shipped day-0 support in transformers, llama.cpp, and vLLM. Meta also promised Muse Spark 1.2 open weights soon.