Midjourney V6, the latest iteration of the popular image generation AI model developed by the research collective founded by David Holz, was released as an alpha version last night. Early users are expressing excitement about the significant improvements it brings. VentureBeat employs Midjourney and various AI art tools for article imagery creation.
Key enhancements in this release include notably improved and more realistic, highly detailed images. Notably, the model now has the capability to generate legible text within images, a feature that had been absent in Midjourney since its initial launch in 2022, while competitors like OpenAI’s DALL-E 3 and Ideogram had already introduced this functionality.
David Holz, in a message posted on the Midjourney Discord server with over 17 million members, highlighted that V6 is the “third model trained from scratch on our AI superclusters” and required nine months of development.
The update does not automatically apply for users. To enable it, users can enter the slash command “/settings” in the Midjourney Discord server or in a direct message to the Midjourney bot. Then, they can use the dropdown menu to select V6. Alternatively, users can manually type “–v 6” after their prompts to activate the update.