After nearly a year-long wait, Midjourney has launched V7, its latest AI image model, which entered alpha testing around midnight EST on Thursday.
To access V7, users must first rate approximately 200 images to create a Midjourney “personalization” profile, which tailors the model to their specific visual preferences and is enabled by default in V7. Once the profile is set, users can activate or deactivate V7 on Midjourney’s website or through its Discord chatbot; the model can be selected from a drop-down menu on the website.
Midjourney CEO David Holz described V7 on X as a “totally different architecture” that is “much smarter with text prompts,” producing noticeably higher image quality with beautiful textures and significantly better coherence in bodies, hands, and objects.
V7 offers two versions, Turbo and Relax, and includes a new Draft Mode, which renders images at ten times the speed and half the cost of the standard mode, although the draft images are of lower quality but can be enhanced and re-rendered.
Holz noted that standard Midjourney features like image upscaling and retexturing are not yet available for V7 but are expected to be added in the near future, possibly within two months.
Midjourney, founded in 2022 by David Holz, expects to generate around $200 million in revenue and is currently establishing a hardware team for unannounced projects while training new video and 3D object generation models.
Midjourney faces lawsuits alleging it infringed on the rights of millions of artists by training its AI tools on images scraped from the web without consent.
