Midjourney has released its first new model in a while, dubbed V7, which is now available in alpha to users. The update focuses on two key areas of improvement: image quality and tools.
In terms of images, V7 promises better coherence and consistency across hands, fingers, bodies, and various objects. It also introduces more detailed and realistic textures and materials, reducing the noticeable “AI-generated” feel that was often associated with previous models.
The new features are centered around the “Draft Mode,” which allows users to generate images at a fraction of the cost and speed compared to other modes. However, Draft Mode comes with lower image quality, making it unsuitable for final products.
V7 includes two main modes: Turbo and Relax. Turbo produces high-quality images quickly but is twice as expensive in terms of credits used. Relax mode takes its time but uses half the credits. A standard mode is currently in development, expected to be refined later.
Source: https://arstechnica.com/ai/2025/04/midjourney-introduces-first-new-image-generation-model-in-over-a-year