Midjourney has been working on v7 of its generative AI tool since the beginning of the year, in parallel with the development of v6. This evolution of the tool could happen quickly.
The next step for Midjourney is already on the horizon. At the end of March, the generative artificial intelligence team began to deliver the first details of its graphics creation tool, as well as a provisional release window. This, while the development of the current version (Midjourney v6) is not completed.
According to the latest information communicated on March 27, there is talk of aiming for a release of Midjourney v7 between one and three months, according to AI consultant Alie Jules. The latest reports shared on the official Discord server also suggest a priority given to the development of Midjourney v7, according to an update made on March 13.
Priority to Midjourney V7, artificial video
The description of the expected improvements remains quite general for the moment: we expect better image quality (even if version 6 of Midjourney already allows photorealistic renderings), but also improved aesthetics. A faster understanding of instructions (prompts) is also mentioned.
The progress updates made by Midjourney in March all emphasized the same point: the leap between V6 and V7 of generative AI will be even greater than the leap between V5 and V6. This is obviously at this stage of communication: it remains to be seen the degree of progress of the tool. This, while Midjourney V6 is not finalized.
Currently, Midjourney V6 is still in alpha version — its release took place in December 2023. A beta version is currently being designed. That said, it appears from Midjourney’s interventions that this V6 is complex, which makes adding features difficult. Moreover, the team rather favors the development of V7, with the intention of simplifying the system to make it evolve more easily.
The development of an artificial video generation service, like what OpenAI’s Sora does, has also been under consideration since February, according to notes taken by the community. The turning point in the video appears linked to V7, but the availability of this functionality could be decoupled from this update, with an arrival by the end of the year.
The clear desire to focus on V7 of generative AI should have the effect of shortening the development of V6 — there could be one final notable update to improve hands, bodies and text rendering, as well as an optimization to accelerate the speed of generating visuals by 20%.
Follow AI news by subscribing to our Artificial newsletter! Written by AI, verified by Numerama.
If you liked this article, you will like the following: don’t miss them by subscribing to Numerama on Google News.