Meta Launches Muse Spark AI Model, Powered by New Superintelligence Team
Key Points
- Meta unveiled Muse Spark, its newest AI model, on Wednesday.
- The model was built by Meta Superintelligence Labs, led by Scale AI co‑founder Alexandr Wang.
- Muse Spark (codenamed Avocado) already powers the Meta AI app and website.
- Within weeks it will extend to WhatsApp, Instagram, Facebook, Messenger, and upcoming AI glasses.
- Meta describes the model as small, fast, and capable of handling complex scientific, math and health queries.
- The launch follows a $14.3 billion investment in Scale AI and a high‑cost talent acquisition strategy.
- Competitors Google and OpenAI have recently released Gemini 3 and GPT‑5 updates, intensifying the AI race.
- Analysts see the move as part of Meta’s broader push toward artificial general intelligence.
Meta unveiled its latest artificial‑intelligence model, Muse Spark, on Wednesday, marking the first product rollout since the company assembled a high‑cost superintelligence team. Built by Meta Superintelligence Labs and led by Scale AI co‑founder Alexandr Wang, the model—internally codenamed Avocado—already runs the Meta AI app and website and will soon power WhatsApp, Instagram, Facebook, Messenger and upcoming AI glasses. Meta describes Muse Spark as a fast, small model capable of tackling complex scientific, mathematical and health queries, while hinting at larger models in the pipeline as it strives to close the gap with rivals like Google and OpenAI.
Meta announced the debut of Muse Spark, its newest artificial‑intelligence model, during a Wednesday press event. The rollout represents the first major AI release since the company forged a $14.3 billion partnership with Scale AI and recruited a cadre of top talent under the banner of Meta Superintelligence Labs.
Led by Alexandr Wang, the co‑founder and CEO of Scale AI, the Superintelligence team was assembled last year from engineers and researchers who previously worked at competing AI firms. Wang’s move to Meta followed the multibillion‑dollar investment that gave Meta a foothold in Scale’s technology and expertise.
Muse Spark, internally referred to as Avocado, is already powering the Meta AI app and the company’s AI‑focused website. Within weeks, the model will expand to the core suite of Meta products—WhatsApp, Instagram, Facebook, Messenger—and even the forthcoming AI glasses that the company has hinted at in recent developer previews.
Meta’s messaging emphasizes the model’s speed and efficiency. "This initial model is small and fast by design, yet capable enough to reason through complex questions in science, math and health," a company spokesperson said. The firm frames Muse Spark as an "early data point on our trajectory," signaling that larger, more powerful models are already in development.
The launch arrives after Meta’s previous Llama series fell short of expectations, prompting criticism that the company lagged behind the rapid advances of its rivals. Google recently unveiled Gemini 3, a model praised for its coding and research capabilities, while OpenAI pushed forward with updates to its upcoming GPT‑5 system.
Industry analysts view Meta’s aggressive investment in AI talent and infrastructure as a strategic push to close the gap with Google and OpenAI. By creating a dedicated "superintelligence" team, Meta signals its ambition to eventually achieve artificial general intelligence, a goal it has publicly pursued for several years.
While Muse Spark is positioned as a foundational model, Meta’s broader AI roadmap suggests a multi‑year plan to scale up model size and capability. The company’s recent hiring spree, combined with the hefty financial commitment to Scale AI, underscores a willingness to pour resources into the AI race, even as the market remains fiercely competitive.
Stakeholders will watch closely how quickly the model can be integrated across Meta’s sprawling ecosystem and whether it can deliver the promised performance gains in real‑world applications. Success could bolster the company’s standing in the AI arena, while any shortcomings may reinforce doubts sparked by the earlier Llama rollout.