{"id":660622,"date":"2025-04-04T08:27:26","date_gmt":"2025-04-04T05:27:26","guid":{"rendered":"https:\/\/en.buradabiliyorum.com\/midjourney-releases-its-first-new-ai-image-model-in-nearly-a-year\/"},"modified":"2025-04-04T08:27:26","modified_gmt":"2025-04-04T05:27:26","slug":"midjourney-releases-its-first-new-ai-image-model-in-nearly-a-year","status":"publish","type":"post","link":"https:\/\/buradabiliyorum.com\/en\/midjourney-releases-its-first-new-ai-image-model-in-nearly-a-year\/","title":{"rendered":"Midjourney releases its first new AI image model in nearly a year"},"content":{"rendered":"
\n

Midjourney, one of the earliest AI image-generating services on the web, has released its first new AI image model in nearly a year.<\/p>\n

Dubbed V7<\/a>, the model, which began rolling out in alpha around midnight Eastern on Friday, comes a week after OpenAI debuted a new image generator in ChatGPT that quickly went viral for its ability to create Ghibli-style photos. Midjourney\u2019s is not a Ghibli-optimized model \u2014 at least not officially \u2014 but nonetheless generates aesthetically pleasing works, at least to this reporter\u2019s dilettante eye.<\/p>\n

To use it, you\u2019ll first have to rate around 200 images to build a Midjourney \u201cpersonalization\u201d profile if you haven\u2019t already. This profile tunes the model to your individual visual preferences; V7 is Midjourney\u2019s first to have personalization switched on by default. <\/p>\n

Once you\u2019ve completed that task, you\u2019ll be greeted with a togglable V7 on Midjourney\u2019s website and, if you\u2019re a member of Midjourney\u2019s Discord server, on its Discord chatbot. In the web app<\/a>, you can quickly select the model from the drop-down menu next to the \u201cversion\u201d label.<\/p>\n

\"Midjourney
Image Credits:<\/strong>Midjourney<\/span><\/figcaption><\/figure>\n

Midjourney CEO David Holz described V7 as a \u201ctotally different architecture\u201d in a post on X<\/a>.<\/p>\n

\u201cV7 is [\u2026] much smarter with text prompts,\u201d Holz continued in an announcement on Discord. \u201c[I]mage prompts look fantastic, image quality is noticeably higher with beautiful textures, and bodies, hands, and objects of all kinds have significantly better coherence on all details.\u201d<\/p>\n

V7 is available in two flavors, Turbo and Relax \u2014 the former of which is more costly to run \u2014 and powers a new tool called Draft Mode that renders images at 10x the speed and half the cost of the standard mode. Draft images are lower-quality than standard-mode images, but can be enhanced and re-rendered with a click of the mouse.<\/p>\n

\n

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<\/a><\/p>\n

\u2014 Midjourney (@midjourney) April 4, 2025<\/a><\/p>\n<\/blockquote>\n

A number of Midjourney capabilities aren\u2019t available yet for V7, according to Holz, including image upscaling and retexturing. Those will arrive in the near future, he said \u2014 possibly as soon as within two months.<\/p>\n

\u201cThis is an entirely new model with unique strengths and probably a few weaknesses\u201d Holz wrote on Discord. \u201c[W]e want to learn from you what it\u2019s good and bad at, but definitely keep in mind it may require different styles of prompting. So play around a bit.\u201d<\/p>\n

\"Midjourney
My prompt: \u201cA dragon with a spiky tail.\u201d<\/span>Image Credits:<\/strong>Midjourney<\/span><\/figcaption><\/figure>\n

In my brief testing, V7 adhered reasonably well to the prompts I gave it. Granted, I didn\u2019t have time to really<\/em> put the model through the ringer.<\/p>\n

\"Midjourney
My prompt: \u201cA Ghibli-style character.\u201d<\/span>Image Credits:<\/strong>Midjourney<\/span><\/figcaption><\/figure>\n

Midjourney is an unusual operation. Started in 2022 by Holz, who co-founded PC peripheral company Leap Motion, it hasn\u2019t taken a dime of outside money.<\/p>\n

In late 2023, Midjourney was reportedly<\/a> expecting to bring in around $200 million in revenue. Recently, the San Francisco-based firm said that it was establishing a hardware team to work on undisclosed projects, and it continues to train previously announced models for video and 3D object generation. <\/p>\n

Several<\/a> lawsuits<\/a> accuse Midjourney of infringing on the rights of millions of artists by training AI tools on images scraped from the web without the consent of the images\u2019 creators.<\/p>\n<\/div>\n