The first new model to come out of Meta Superintelligence Lab following the company’s reorganization of its AI efforts, Muse Spark reflects a shift toward efficient, product-ready AI as enterprises weigh cost, latency, and real-world deployment.

InfoWorld is a source of news, analysis, and commentary on technology trends, IT strategies, and business innovation. With a focus on enterprise technology and digital transformation, InfoWorld offers insights and guidance for IT decision-makers, software developers, and technology professionals. From  articles on cloud computing and cybersecurity to product reviews and industry trends, InfoWorld helps readers navigate the complexities of modern IT environments and make informed decisions to drive business success.

InfoWorld

Meta has introduced Muse Spark, a small and fast AI model developed by its newly reorganized Meta Superintelligence Lab. The model currently powers the Meta AI assistant on the web and in the Meta AI app, with plans to expand to WhatsApp, Instagram, Facebook, Messenger, and Meta's smart glasses. An API will be offered to select partners in private preview, with open-sourcing of future versions planned. Muse Spark supports multimodal inputs, multiple reasoning modes, and parallel sub-agents for complex queries, making it suited for enterprise use cases like customer support, automation, and internal copilots. Meta published results across 20 AI benchmarks, showing competitive but not dominant performance against rivals like Claude, Gemini, and GPT. The model received targeted training for health-related queries and underwent extensive safety evaluations.

Meta’s Muse Spark: a smaller, faster AI model for broad app deployment