Midjourney launches new V7 imaging model: what's new?
Midjourney has released the alpha version of their new V7 image generation model, which dramatically improves text query processing. The company claims it's their smartest, most beautiful, and most holistic model yet.
Here's What We Know
The new model features enhanced image quality, improved textures and more accurate reproduction of details such as hands and bodies. This is an important achievement, as AI-based image generators often struggle to create realistic images of hands.
We're now beginning the alpha-test phase of our new V7 image model. It's our smartest, most beautiful, most coherent model yet. Give it a shot and expect updates every week or two for the next two months. pic.twitter.com/Ogqt0fgiY7
- Midjourney (@midjourney) April 4, 2025
One of the key features of the V7 is "Draft Mode," which allows you to create images ten times faster and for half the cost of standard mode. Although the quality of these images is lower, they can be enhanced and reworked to produce the final result. The model also supports Turbo and Relax modes for standard rendering, where the former provides faster performance but costs two times more.
To test V7, users need to activate personalisation, which trains the model for visual preferences. This is achieved by ranking 200 pairs of images. Personalisation is enabled by default, but it can be turned off. Midjourney plans to release updates for V7 every one to two weeks over the next two months.
Source: @midjourney, engadget
