Midjourney, previously usable only via Discord, launched a website that lets users generate images directly from a browser. New users can also try it for free.
Midjourney has released V6, a model developed over roughly nine months, which can now render text within images and interprets prompts in a substantially different way.
Midjourney released an alpha version of its V5 model. Output resolution doubled, and hand depiction โ long a weak point of AI image generation โ improved dramatically.
Midjourney released its V4 model, built on a completely new AI architecture. Depicting people and objects became far more accurate, and it became the foundation for all versions that followed.
Midjourney ended its invite-only closed beta and moved to an open beta anyone could join simply by joining its Discord server, an event that became the starting point of the image generation AI boom.