Midjourney has released Version 5.2 for testing. According to the developers, this new version is more advanced and capable of generating realistic images.
The tool’s understanding of prompts has been improved. It is more consistent to analyze what is written than before. The statement states, “This model produces more detailed, sharper results with better colors, contrast, and compositions. It also has a slightly better understanding of prompts than earlier models and is more responsive to the full range of the –stylize parameter.”
One of the most important new features in Midjourney Version 5.2 is the new Shorten command. This command analyzes typed texts to create images. It finds out which words can work and can be used as keywords, or vice versa, which words have no effect, and offers suggestions. Currently, version 5.2 is used by default; users can revert to the previous version 5.1.
Thus, the new v5.2 has a Zoom-out feature that can create image versions with different camera angles. In addition, the artificial intelligence tool automatically fills in the missing parts of the image while zooming out.
Join the Artificial Intelligence Bundle workshop to unlock the secrets of prompt crafting with a diverse group of instructors. In Carlos Banon’s Taking Control 2.0: Midjourney x ControlNet, Shail Patel’s Prompt Intelligence, and Hassan Ragab’s AI Conceptual Architecture 3.0 will comprehensively explore new techniques in Midjourney, focusing on different architecture and design concepts.