{"slug": "muse-glimmer-from-meta-superintelligence-labs-is-now-available", "title": "Muse Glimmer from Meta Superintelligence Labs is now available", "summary": "Meta Superintelligence Labs released Muse Glimmer, a 30B multimodal open model for local agent workloads, now available on Ollama with initial support for Apple Silicon via the MLX engine. The model, released under Apache 2.0, features a 128K+ context length and a 1.8B-parameter perception encoder for image understanding, and runs 1.5x–1.8x faster with DFlash support. It can power coding agents like Claude Code and Codex, as well as personal assistants such as OpenClaw and Hermes.", "body_md": "# Muse Glimmer from Meta Superintelligence Labs is now available\n\n## August 10, 2026\n\nNote: Muse Glimmer is currently available via initial support via Ollama’s MLX engine on Apple Silicon. Additional support and optimizations for Apple Silicon, NVIDIA, AMD, and other platforms will be available in the coming days.\n\n**Muse Glimmer**, Meta’s newest open model and the first released by Meta Superintelligence Labs, [is now available on Ollama](https://ollama.com/library/muse-glimmer). It’s a 30B multimodal model purpose-built for agent workloads that run locally with a 128K+ context length, released under the Apache 2.0 license.\n\nWith Ollama, you can now use Muse Glimmer to power coding agent applications such as Claude Code, Codex, Pi and more, as well as long-running personal assistants such as OpenClaw and Hermes.\n\nOllama’s MLX engine provides state-of-the-art performance on Apple Silicon for this model, with support for DFlash and image input.\n\nTo run Muse Glimmer locally, [download the latest release of Ollama](https://ollama.com/download) and run:\n\n```\nollama run muse-glimmer:30b-mlx\n```\n\n## Power coding agents locally\n\nTo run Muse Glimmer on Apple Silicon with Claude Code, [download Ollama](https://ollama.com/download) and run:\n\n```\nollama launch claude --model muse-glimmer:30b-mlx\n```\n\nFor a lighter-weight coding agent, try [Pi](https://pi.dev):\n\n```\nollama launch pi --model muse-glimmer:30b-mlx\n```\n\nFor personal assistant frameworks such as OpenClaw and Hermes, use:\n\n```\nollama launch openclaw --model muse-glimmer:30b-mlx\nollama launch hermes --model muse-glimmer:30b-mlx\n```\n\n`ollama launch`\n\nalso works with Codex, OpenCode, GitHub Copilot, and more. See [all integrations](https://docs.ollama.com/integrations).\n\nMuse Glimmer also supports controllable reasoning strength: `low`\n\n, `medium`\n\n, `high`\n\n, and `xhigh`\n\n. Use `high`\n\nor `xhigh`\n\nfor complex coding and agentic tasks, and lower strengths when speed matters more.\n\n## State-of-the-art performance on Apple Silicon with new DFlash and image input support\n\nOllama’s [MLX engine](/blog/mlx-performance) now supports DFlash, building on previous [multi-token prediction (MTP) support](/blog/faster-gemma-4-mlx-mtp). With DFlash, Muse Glimmer runs 1.5×–1.8× faster on Apple Silicon.\n\nMuse Glimmer’s dedicated 1.8B-parameter perception encoder gives it native image understanding. With new image input support in the MLX engine, coding agents can use it for low-latency, back-to-back tool calling on tasks that require images, such as:\n\n- Building websites or applications from a drawing or mockup\n- Computer use applications powered by screenshots\n- Reading documents, receipts, and charts", "url": "https://wpnews.pro/news/muse-glimmer-from-meta-superintelligence-labs-is-now-available", "canonical_source": "https://ollama.com/blog/muse-glimmer", "published_at": "2026-08-10 19:23:23+00:00", "updated_at": "2026-08-10 19:42:00.664760+00:00", "lang": "en", "topics": ["artificial-intelligence", "large-language-models", "generative-ai", "ai-products", "ai-infrastructure"], "entities": ["Meta Superintelligence Labs", "Muse Glimmer", "Ollama", "MLX", "Apple Silicon", "Claude Code", "Codex", "OpenClaw"], "alternates": {"html": "https://wpnews.pro/news/muse-glimmer-from-meta-superintelligence-labs-is-now-available", "markdown": "https://wpnews.pro/news/muse-glimmer-from-meta-superintelligence-labs-is-now-available.md", "text": "https://wpnews.pro/news/muse-glimmer-from-meta-superintelligence-labs-is-now-available.txt", "jsonld": "https://wpnews.pro/news/muse-glimmer-from-meta-superintelligence-labs-is-now-available.jsonld"}}