Midjourney launched its newest image-generation synthetic intelligence (AI) mannequin, V7, on Friday. The V7 AI mannequin is at present obtainable in an alpha model, and the corporate is at present letting its consumer base check the mannequin’s functionality. Arriving practically a yr after its predecessor V6, the V7 Alpha mannequin can also be the primary within the AI agency’s lineup to have personalisation turned on by default. The basis mannequin additionally introduces a brand new “Draft Mode,” which is claimed to value half as a lot as the usual mode whereas providing 10 occasions the velocity.
Midjourney’s V7 Brings New Modes and Features
In a publish on X (previously often called Twitter), the official deal with of Midjourney introduced the discharge of V7. Compared to V6, the picture era mannequin has the next textual content immediate adherence, and each the picture high quality and textures are higher, the corporate added. The AI agency additionally claimed that the coherence amongst totally different parts is increased than its predecessor.
Midjourney highlighted that V7 is its first mannequin to have mannequin personalisation turned on by default. It signifies that earlier than gaining access to the AI mannequin, customers must fee a lot of pictures to construct their personalisation profile. This profile permits the Midjourney fashions to grasp the consumer’s desire, and it might probably regulate the outputs accordingly. However, as soon as a consumer will get entry to V7, they’ll flip personalisation on or off.
The firm can also be introducing a brand new function dubbed Draft Mode with this mannequin. Draft Mode is claimed to be half the price of the usual mode and renders pictures at 10 occasions its velocity. It is basically an iteration mode the place customers may give the mannequin a number of prompts to get the specified output and to make inline edits to photographs. The mode helps voice enter for a conversational expertise.
One of probably the most thrilling new options for our new V7 mannequin is one thing we name “Draft Mode”. Draft mode is half the associated fee and 10 occasions the velocity and it is likely to be one of the simplest ways to iterate on concepts ever. Try it with voice, suppose out loud and let our concepts move like liquid desires. pic.twitter.com/ANfTMC6Ej1
— Midjourney (@midjourney) April 4, 2025
Users can both click on “Draft mode” on the internet consumer or add “–draft” on the finish of the immediate to activate the mode. Midjourney highlighted that draft mode pictures are decrease high quality than the usual mode, however the mannequin behaviour and aesthetic era are constant.
The V7 picture era mannequin is offered in Turbo and Relax variants. Notably, Turbo mode prices twice as a lot as commonplace output era. Additionally, the V7 mannequin at present doesn’t help upscaling, modifying, and retexture-related duties, and it redirects to V6 to finish these duties.
Midjourney additionally plans to launch new options for the AI mannequin each week or two for the subsequent 60 days. One function within the roadmap contains character and object reference.
