Midjourney Unveils Its Most Advanced Image-Generating AI Model V7

One of the pioneering AI image generation services, Midjourney, has unveiled its new AI model, V7.

«This is our most intelligent, aesthetically pleasing, and cohesive model,» the announcement states.

V7 is available in two versions: Relax and the more expensive Turbo option. Additionally, there’s a Draft Mode, which is half the price and ten times faster than the standard plan. While draft images are of lower quality, they can be enhanced immediately. The company encourages users to «think aloud,» verbalizing various ideas for image generation.

Before starting with V7, users must provide the neural network with around 200 images to create a personalized profile. This profile tailors the model according to the user’s individual preferences.

«Sometimes you need to experiment a bit with the prompt—this has a completely different architecture,» emphasized CEO David Holz.

Several standard features of Midjourney, such as image scaling and retexturing, are currently unavailable for V7 but will be added in the future.

The latest development from Midjourney was introduced over a year ago, when the company announced a new AI model for image generation capable of producing coherent text.

This new tool emerged a week after OpenAI integrated an image generator based on the GPT-4o model into both ChatGPT and Sora. The service gained significant popularity, leading to a shortage of processing power.

Just a reminder, in April, Sam Altman’s startup secured a new funding round, raising $40 billion at a valuation of $300 billion.