Meta launches Muse Spark 1.1, an advanced multimodal reasoning model

Jul 9, 2026

AI Summary

Meta has introduced Muse Spark 1.1, a significant upgrade designed for agentic tasks with enhanced capabilities in coding, tool use, and multimodal understanding. The model is now available for developers through the new Meta Model API, aiming to improve efficiency in complex workflows.

Muse Spark 1.1 is a multimodal reasoning model developed by Meta Superintelligence Labs, designed for agentic tasks.
The model shows improvements in tool and computer use, coding, and multimodal understanding compared to its predecessor, Muse Spark.
It can perform personal agentic tasks, manage context effectively, and orchestrate multi-agent systems to optimize performance.
Muse Spark 1.1 maintains context across multiple applications and adapts to changing requirements with minimal human intervention.
The model has demonstrated significant improvements in coding tasks, including diagnosing bugs and implementing features in complex codebases.
It excels in multimodal capabilities, allowing it to interact with real environments and generate outputs based on visual and audio inputs.
Extensive safety evaluations indicate that Muse Spark 1.1 operates within safe margins and shows resilience against various threats.
Developers can access Muse Spark 1.1 through the new Meta Model API, which is currently in public preview, with early partners praising its capabilities for agentic workloads.

musemetaai toolsproduct updateevaluation

Meta launches Muse Spark 1.1, an advanced multimodal reasoning model

Related Stories

OpenAI discontinues Atlas but enhances AI browsing features in ChatGPT and Chrome extension

Meta launches Muse Spark 1.1, a new AI coding model to compete in the market

Meta introduces AI image generator, raises concerns over use of public Instagram photos

Meta to Begin Production of New AI Chips in September to Reduce GPU Costs