Meta releases Muse Spark Open: most capable US open-weight model, runs on two Mac Studios
AIatMeta · x · 2026-10-16
Meta has released Muse Spark Open, billed as the most capable US open-weight model to date: a natively multimodal 816B-parameter MoE (42B active) with a 1M-token context window, built for coding and agentic work — long-horizon tasks, complex software engineering, and multi-step reasoning. Notably, it's the first model in its class to run on a desktop: two Mac Studios with M5 Ultra hold the entire model via the latest MLX framework (up to 512GB unified memory each). Weights ship under the Muse Model License with limited commercial terms allowing most developers free use. Available via Baseten, Fireworks AI, Groq, Modal, Nebius, OpenRouter, and Together AI; downloadable from SageMaker, Google Cloud Model Garden, Hugging Face, and Microsoft Foundry; runs on SGLang, vLLM, TensorRT-LLM, llama.cpp, Unsloth, and more, optimized for AMD, Apple, Intel, and NVIDIA silicon.
More from Infra
- Meta's 816B open-weight Muse Spark Open hits Groq preview at 350+ tokens/sec — GroqLLC · 2026-10-16
- Together AI serves Muse Spark Open serverless for agentic coding and multimodal workloads — togethercompute · 2026-10-16
- Musk says SpaceX-Nvidia 'Star Mind' space datacenter computer hits volume production next year — elonmusk · 2026-10-16
- Muse Spark Open lands on MLX just 30 minutes after release — pcuenq · 2026-10-16
- TSMC earnings note cross-referenced with WFE model and power checks in CS Atlas — BenBajarin · 2026-10-16
- Vercel Drives is GA: persistent storage lets stateful agents survive environment changes — evilrabbit_ · 2026-10-16