Midjourney is released V7, the first new AI image model in almost a year

Midjourney, one of the first AI -based image generators, has released the first new AI image model in almost a year.

Compiler V7The model began to be released on Alpha around midnight Ests on Thursday, a week after the openai debut of a new image generator in Chatgpt who quickly became a viral for his ability to create Ghibli type photos.

Midjourney’s is not a model that is optimized with Ghibli-at least not officially-but can create aesthetically pleasing works, at least in the journalist’s dilettante eye.

We are now starting the Alpha-Test phase of the new V7 image model. It is our smartest, most beautiful, most cohesive model yet. Give him a shot and expect updates each week or two for the next two months. pic.twitter.com/ogqt0fgiy7

– Midjourney (@Midjourney) April 4 2025

To use it, you must first evaluate about 200 images to create a “personalization” Midjourney profile if you don’t already have it. This profile adjusts the model to your individual visual preferences. The V7 is the first Midjourney model to have the default personalization to be activated.

Once you do this, you will be able to turn on or turn off the V7 on the Midjourney website and, if you are a member of the Midjourney Discord server in Discord Chatbot. In the web application, you can quickly select the model from the drop -down menu next to the “version” label.

Midjourney CEO David Holz described V7 as “completely different architecture” in a Post in x.

‘V7 is […] Much smarter with the text prompts, “Holz continued in a statement about the disagreement.”[I]Mage’s prompts look fantastic, image quality is significantly higher with beautiful textures, and the bodies, hands and objects of all kinds are significantly better in all details. “

The V7 is available in two flavors, turbo (expensive to run) and relax and authorize a new tool called Draft Mode that makes images at 10x speed and half of the cost of standard operation. The images drawing is of lower quality than standard functional images, but can be reinforced and repeated with one click.

Some standard Midjourney features are not yet available for the V7, according to Holz, including upgrading image and review. These will reach the near future, he said, possibly within two months.

“This is a completely new model with unique advantages and possibly some weaknesses,” Holz wrote on Discord. “[W]You want to know from you what is good and bad, but you definitely keep in mind that it can require different styling styles. So play a little. ”

In my short trials, V7 has meant pretty well in my prompts, though I didn’t have time to really Put the model through the bell.

Midjourney is an unusual operation. He started in 2022 by Holz, who founded the PC Peripheral Company Leap Motion, did not set a penny of external money.

In late 2023, Midjourney was referenced Expecting to bring about $ 200 million in revenue. Recently, the San Francisco -based company said it has created a material team to work in some projects that do not describe in detail and continues to train previously announced models for video production and 3D.

The company is facing several lawsuits This accuses him of violating the rights of millions of artists by drawing up AI tools in images scratching from the tissue without the consent of the creators of the images.