Meta Launches Muse Image and Muse Video Models for Media Generation

On July 7, 2026, Meta Superintelligence Labs launched Muse Image, an advanced image generation model, and previewed Muse Video, a media generation model. Muse Image currently ranks No. 2 on the Arena Elo rankings for text-to-image, single-image editing, and multi-image editing as of July 5, 2026. Muse Video holds the No. 3 position in human-preference Elo for text-to-video.

Muse Image is accessible through the Meta AI app, on meta.ai, Instagram Stories in the US, and WhatsApp in select countries. Meta claims that Muse Image is their most advanced image generation model to date, featuring a hidden provenance signal for images created within the Meta AI ecosystem. Muse Video is designed to offer competitive performance in prompt adherence, visual fidelity, and temporal consistency. This launch follows Meta's recent introduction of AI agents for WhatsApp business setup and management, indicating a broader strategy to integrate advanced AI capabilities across its platforms.