Meta releases Muse Spark Open with 816 billion parameters

Oct 15, 2026
Meta released Muse Spark Open, a multimodal model for coding, reasoning and agent tasks that can run locally on two Apple M5 Ultra Mac Studios.
Meta releases Muse Spark Open with 816 billion parameters

Meta Platforms, Inc. released the weights for Muse Spark Open, the company said in an announcement. The multimodal mixture of experts model has 816 billion total parameters, activates 42 billion parameters per token and supports a context window of 1 million tokens.

Muse Spark Open is designed for coding, tool use, extended reasoning and agent tasks. It accepts text, images, video and audio as input and produces text. Meta says the model can run locally on two Apple M5 Ultra Mac Studios using MLX. Other local configurations require at least 700 GB of GPU memory for the NVFP4 version or 2 TB for BF16.

The weights are available through Hugging Face, Amazon SageMaker AI, Google Cloud Model Garden and Microsoft Foundry Models. API access is available from Baseten, Fireworks AI, Groq, Modal, Nebius, OpenRouter and Together AI.

Meta released the model under the Muse Model License. Businesses with more than $20 million in aggregate gross revenue over any consecutive 12 month period must reach a separate agreement with Meta before offering the model through a model as a service business.

We hope you enjoyed this article

Consider subscribing to one of our newsletters like Daily AI Brief.

Also, consider following us on social media:

AI Brief AI Brief (X)
Free newsletter

Daily AI Brief

Daily report covering major AI developments and industry news, with both top stories and complete market updates

AI Token Webinar

Whitepaper

Tensordyne Napier: What If One Rack Could Do the Work of Nine?

Tensordyne

This Tensordyne whitepaper presents Napier, an inference-focused AI processor and rack-scale system based on the company’s TDN Math logarithmic number system. It examines infrastructure requirements for large mixture-of-experts and agentic models, compares major inference architecture approaches, and details the TDN AIP processor, TDN72 pod, TDN Link fabric, and Napier Ultra configuration. The paper reports simulation-based performance, cost, and accuracy-validation results, including Tensordyne’s projected comparison of one Napier rack with a nine-rack Nvidia Rubin plus Groq deployment; the chip is reported as taped out and in fabrication.

Read more

You may also like

Free, six days a week

Daily AI Brief: the AI news that matters, in your inbox.