Midjourney Unveils V7 AI Image Model with Personalization Features

Midjourney has released its latest AI image model, V7, marking its first new model in nearly a year. The model, which began rolling out in alpha, introduces personalization features that allow users to tailor the model to their visual preferences by rating around 200 images. This personalization is a first for Midjourney, enhancing the model's ability to generate images that align with individual tastes.

V7 is described as having a "totally different architecture" and offers improved handling of text and image prompts, resulting in higher image quality with richer textures and better coherence in details such as bodies and hands. The model is available in two modes: Turbo, which is costlier, and Relax. Additionally, V7 powers a new Draft Mode that renders images at ten times the speed and half the cost of the standard mode, although these draft images are of lower quality and can be enhanced later.

Some features, such as image upscaling and retexturing, are not yet available for V7 but are expected to be introduced within the next two months. Midjourney continues to engage with its community through platforms like Discord, where users can provide feedback and explore the model's capabilities.