Meta has announced the launch of Muse Spark, the first model in its new Muse family developed by Meta Superintelligence Labs, marking a significant step in the company’s push toward personal superintelligence. The newly introduced system is a natively multimodal reasoning model designed to handle complex tasks across text, images, and tools, while enabling advanced capabilities such as visual chain-of-thought reasoning and multi-agent orchestration.

Muse Spark represents the first outcome of a broader overhaul of Meta’s AI strategy, with the company making substantial investments across research, model training, and infrastructure. This includes the development of the Hyperion data centre, which is expected to support the next phase of large-scale AI model training and deployment.

The model delivers competitive performance across a range of domains, including multimodal perception, reasoning, healthcare-related tasks, and agentic workflows. Meta also introduced a new “Contemplating mode,” which orchestrates multiple AI agents to reason in parallel, enhancing performance on complex, multi-step problems.

This feature enables Muse Spark to compete with advanced reasoning systems such as Gemini Deep Think and GPT Pro. According to Meta, the model achieved 58 per cent on Humanity’s Last Exam benchmark and 38 per cent on FrontierScience Research tasks, reflecting improvements in handling challenging reasoning scenarios.

Muse Spark is described as the first step in Meta’s “scaling ladder,” with larger and more capable models already in development. The company noted ongoing efforts to address current limitations in areas such as long-horizon agentic systems and coding workflows.

The model is now available via meta.ai and the Meta AI app, with a private API preview being rolled out to select users. As Meta continues to expand its AI stack, Muse Spark signals a strategic shift toward more integrated, agent-driven systems aimed at delivering highly personalized and context-aware intelligence.

With this launch, Meta is positioning itself more aggressively in the race to build next-generation AI systems capable of advanced reasoning and real-world task execution at scale.