The initial model is capable of reasoning through complex questions in science, mathematics, and health, the company says.
Meta Platforms on Wednesday introduced Muse Spark, its first artificial intelligence model developed by a high-cost team formed last year to help the company compete in the fast-moving AI race.
Following the announcement, the company’s shares rose further, extending gains to nearly 7%.
US tech giants are facing growing pressure to demonstrate that their enormous investments in artificial intelligence will deliver meaningful returns. The pressure is particularly intense for Meta, which last year brought in Scale AI CEO Alex Wang under a $14.3 billion deal and reportedly offered some engineers compensation packages worth hundreds of millions of dollars to build a new superintelligence team.
Superintelligence refers to AI systems that could surpass human intelligence. Muse Spark is the first model released from this group and belongs to an internally developed family of models known as Avocado.
The model will initially be accessible only through the lightly used Meta AI app and website. In the coming weeks, it is expected to replace existing Llama models that currently power chatbots across WhatsApp, Instagram, Facebook, and Meta’s smart glasses lineup.
“In its initial form, the model is small and fast by design, yet capable of reasoning through complex questions in science, math, and health. It provides a strong foundation, with the next generation already in development,” the company said in a blog post.
Meta did not reveal the model’s size, a key metric typically used to compare AI systems’ computing power with competitors.
Meta CEO Mark Zuckerberg had previously tempered expectations, telling investors in January that the first models from the team would “be good but, more importantly, will show the rapid trajectory that we’re on.”
“I expect us to steadily push the frontier over the course of the year as we continue to release new models,” he said.
Focus on everyday tasks
Muse Spark is designed to assist with practical tasks, such as estimating the calories in a meal from a photo or placing an image of a mug onto a shelf to visualize how it would look.
Meta has also introduced a feature called Contemplating mode, which uses multiple AI agents simultaneously to improve reasoning ability. For example, when planning a family vacation, one agent can create an itinerary while another searches for kid-friendly activities.
The company is betting that AI-driven everyday assistance will increase engagement across its platforms, which serve more than 3.5 billion users, potentially giving Meta an advantage over competitors with smaller user bases.
About the Author
