Midjourney released the alpha version of its new AI model, V7, on Thursday. It will have to compete with the image generator in ChatGPT, which went viral for its ability to create Ghibli-style photos. This is Midjourney’s first release since December 2023, when V6 was released .
Midjourney released the alpha version of its new AI model, V7, on Thursday. It will have to compete with the image generator in ChatGPT, which went viral for its ability to create Ghibli-style photos. This is Midjourney’s first release since December 2023, when V6 was released .
To use V7, you first need to evaluate about 200 images to create a Midjourney “personalization” profile, if one hasn’t been created before. This profile customizes the model to your individual visual preferences; V7 is the first Midjourney model to have personalization enabled by default.
“The V7 is the first model to have model personalization enabled by default. To use it, you need to unlock personalization. It takes about 5 minutes. You can turn it on/off at any time. We believe personalization raises the bar for how well we can interpret what you want and what you find beautiful,” explains Midjourney.
Once you complete this task, you will see the V7 version you can switch to on the Midjourney website and, if you are a member of the Midjourney Discord server, in the Discord chatbot. In the web app, you can quickly select the model from the drop-down menu next to the "version" label.
"V7 is much smarter with text clues," said Midjourney CEO David Holtz on Discord. "The image clues look fantastic, the image quality is noticeably higher with beautiful textures, and bodies, hands, and all sorts of objects have much better coherence of all the details."
V7 comes in two flavors, Turbo and Relax — the former of which is more expensive to use — and uses a new tool called Draft Mode , which renders images 10 times faster and at half the cost of standard mode. Draft images are lower quality than standard mode images, but they can be enhanced and re-rendered with a single click, TechCrunch reports .
According to Holtz, some Midjourney features are not yet available in V7, including image scaling and retexturing. He says they will be available in the near future — possibly within two months.
As a reminder, OpenAI's new image generator based on the GPT-4o model is now available to all users .
