Rival for Chatgpt: Midjourney launched its new model generation model after the huge success of the “Ghibli style”

Midjourney, one of the first artificial intelligence-based image generators, has launched its first new images model after almost a year, Techcrunch reports.

Called V7, the model began to be implemented in the Alfa version of Midjourney on Thursday around the middle of the night and comes a week after Openai launched a new image generator in Chatgpt, which has quickly become viral due to its ability to create images in the style of the renowned Ghibli Studio in Japan.

The Midjourney model is not officially optimized for the Ghibli style, but it can generate impressive aesthetic works, as the examples published by the company shows and confirms Techcrunch.

Those who have not used Midjourney will first have to evaluate about 200 images to build a “personalization” profile for AI AI profile adjusts the model according to users' visual preferences and V7 is the first Midjourney model that has the customization activated by default.

Midjourney V7 – My First Images

They is All Straight Out of V7. No upscaling.

First Impression is that the – personsonization plays a heavy part in how your images look. My Images has Very Film-Like, with a Mythy Greenish Tones in it.

Still Trying to Figure out if there is a… pic.twitter.com/vzslewgers

– James Yeung (@Jamesyeung18) April 4, 2025

After completing this step, the V7 can be activated or disabled on the Midjourney site or the chatbot on the discord of the platform, for those who are part of the group.

Midjourney V7 V7 is described by the company as “completely new”

CEO Midjourney, David Holz, described V7 as a “completely different architecture”.

“V7 is […] much smarter in interpreting textual commands. The commands for the images look fantastic, the quality of the images is visibly higher, with beautiful texture, and the bodies, hands and objects of any type have a significantly better coherence in all details, ”he said in a message published on the Midjourney Discounted Channel.

V7 is available in two variants – Relax and Turbo (more expensive to use) – and feeds a new tool called Draft Mode, which generates 10 times faster and half of the standard mode. The images generated in draft mode are of lower quality compared to those in standard mode, but can be improved and regenerated by a single click.

One of the Most Exciting New Features for Our New V7 Model is Something We Call “Draft Mode”. Draft mode is Half the cost and 10 times the special and it might be the best way to itrates on ideas. Try it with Voice, Think out Loud and Let Our Ideas Flow Like Liquid Dreams. pic.twitter.com/anftmc6ej1

– Midjourney (@midjourney) April 4, 2025

According to Holz, some of the standard functions of Midjourney are not yet available for V7, including “upscaling” and image retext. These will be added in the near future, possibly within two months.

“This is a completely new model, with unique strengths and probably a few weaknesses,” Holz wrote on the discord. “We want to learn from you what does well and where it has problems, but keep in mind that it may require different command styles. So experience a little,” he urged.