Run Meta's open-weight Muse Spark Open yourself
Meta released Muse Spark Open on October 15 and calls it the most capable US open-weight model to date: 816B total and 42B active parameters, with a 1M-token context. It is built for coding and agent work, runs on two Apple M5 Ultra Mac Studios with MLX, and is available on Baseten, Fireworks AI, Groq, Modal, Nebius, OpenRouter and Together AI.
Why it matters
A large open-weight model with 1M context gives teams a frontier-class option they can host and control themselves.
Who it's for
ML engineers, startups that need self-hosting, and open-model fans.
How to try it
- Fastest route: call it through OpenRouter or Together AI.
- To self-host, download the weights from Hugging Face and serve them with vLLM, SGLang or llama.cpp.
- Check the license terms in Meta's announcement before commercial use.
Plan & availability: Weights on Hugging Face; hosted on Baseten, Fireworks AI, Groq, Modal, Nebius, OpenRouter and Together AI. Check the license.
Source: @AIatMeta on X ↗
Written by AgentStackList from the original post. Checked Oct 17, 2026. Features, plans and availability can change, so confirm on the source before you rely on it.
open weightsself-hosting
More Muse tricks
- Ask your Muse for a costume idea based on your Instagram likes · Oct 15
- Use Muse Wonder inside Cursor · Oct 15
- Run Meta's Muse Spark Open locally on a big Mac · Oct 15
- Meta publishes 10 health research papers written with Muse models · Oct 14
Agent Stack Weekly · free
Get the week's best AI tricks
Limit and pricing changes, new MCP servers, and setup fixes for Claude Code, Codex, Cursor, Copilot and Antigravity.
No spam. Unsubscribe anytime. Read past issues