OllamaFrom the builder
Muse Glimmer from Meta Superintelligence Labs is now available
Muse Glimmer, Meta's newest open model and the first from Meta Superintelligence Labs, is now on Ollama as a 30B multimodal model for local agent workloads with 128K+ context under the Apache 2.0 license. On Apple Silicon, Ollama's MLX engine adds DFlash support that runs Muse Glimmer 1.5x to 1.8x faster, along with image input. The model has a dedicated 1.8B-parameter perception encoder and offers reasoning strength settings of low, medium, high and xhigh.
With DFlash, Muse Glimmer runs 1.5×–1.8× faster on Apple Silicon.
Saved . A shortened excerpt from the official blog; the original publisher retains authorship.