Muse Image now available on AI Gateway

Muse Image, developed by Meta Superintelligence Labs, has been integrated into Vercel’s AI Gateway as the platform’s first image model. Unlike text-based models, Muse Image processes prompts to produce images or modifies existing images based on instructions. The model supports both generation and editing within a single interface, eliminating the need to switch between different tools for distinct tasks.

Users can refine outputs by providing reference images alongside text prompts, allowing the model to blend styles or elements into the generated result. For editing, users submit an image with a specific instruction, and the model applies the requested changes while preserving unrelated parts of the image. This dual functionality streamlines workflows for creators and developers working with visual content.

AI Gateway serves as a unified API layer for accessing Muse Image and other models, offering features such as usage tracking, cost management, failover mechanisms, and performance optimizations to enhance reliability. The platform ensures higher uptime than individual providers by routing requests efficiently and maintaining redundancy across endpoints.

Pricing for Muse Image aligns with Meta’s provider rates, with no additional markup or platform fees applied to inference costs, including requests made through Bring Your Own Key (BYOK). The AI Gateway interface provides visibility into all available image models, enabling users to compare options and select the most suitable tool for their needs.