Midjourney has opened its AI image generation platform to everyone, allowing users to create up to 25 images freely. The move eliminates the previous requirement of using Discord to craft and edit images. Midjourney's decision was announced by CEO David Holz on Discord.
User Experience Overhaul
Previously, users had to handle generating images through Discord, which posed difficulties for some. To enhance user accessibility, Midjourney launched a dedicated website late last year, initially for those who had already created 10,000 images via Discord. The platform is now available to all, providing a simplified and user-friendly environment.
Users can sign up with a Google or Discord account, with an option for those switching from Discord to view their image history. The website features a sidebar, including sections for browsing others' creations, generating new images, and organizing personal galleries.
The Create section offers a quick tutorial video to guide users. When a prompt is entered, the system yields four image options, which can be further tweaked through various actions like Strong or Subtle effects, zooming, panning, and upscaling. Users can also make additional adjustments via an image editor, allowing changes to the aspect ratio and the inclusion of new elements.
Community Interaction and Subscription Models
The platform houses chat rooms where users can share images, tips, and interact. After the initial free allowance of 25 images, users can opt for paid plans—Basic, Standard, Pro, and Mega—with varying features and image quotas.
Midjourney had earlier suspended free usage after social media saw a surge in AI-generated images created via its service. The latest move to broaden access might be a strategic response to the launch of Ideogram 2.0, a rival AI image generator.
In March, Midjourney introduced enhancements to its AI image generation capabilities, adding character consistency. The character reference (cref) function works by allowing users to generate characters that maintain facial features, body types, and clothing across various prompts. This is achieved by appending the “–cref” tag to the end of a text prompt along with a URL of a previously generated image of the character.
