Meta has officially launched Muse Spark, its most advanced AI model to date and the inaugural release from the Meta Superintelligence Labs (MSL). Engineered to power Meta AI across the company’s entire ecosystem, the model is designed to deliver faster, more accurate, and hyper-personalized experiences for users discovering information and interacting with content.
Muse Spark is purpose-built for Meta’s ecosystem, enabling a smarter and faster Meta AI while laying the foundation for new experiences over time. It draws from content people share across Instagram, Facebook, and Threads to deliver more relevant recommendations, supporting a wider range of everyday needs. Muse Spark currently powers the Meta AI app and website, and will continue expanding across Meta platforms globally, with plans to explore API access for select partners.
Key highlights of Muse Spark include:
- Smarter, more capable Meta AI: Enhances Meta AI’s ability to handle both quick questions and more complex tasks, with stronger reasoning and deeper understanding across text and visual inputs.
- Multimodal understanding: Better interprets what users see and experience—not just what they type. For example, users can snap a photo of a snack shelf and Meta AI can identify and rank options, such as highlighting those with higher protein, or scan a product and ask how it compares to alternatives—helping users make more informed decisions in real-world contexts.
- More personalized experiences: Delivers more relevant and tailored responses by drawing from user preferences, past interactions, and content shared across Instagram, Facebook, and Threads.
- Creative and interactive capabilities: Enables a range of use cases, including visual coding to create custom websites and interactive experiences directly from a prompt. For example, users can ask Meta AI to build a dashboard for planning a big surprise party, create a retro-style mini-game, or generate other interactive experiences—and share them with friends.
- Built for continuous improvement: Designed for scalability, enabling faster iteration and ongoing model improvements.
Google deepens commitment to keeping Thais safer online with expanded initiatives
Companion pet robots offer emotional support for elderly and kids
