1. Midjourney updated random style feature works with current AI image generation model
2. Users can randomize image style by adding “–sref random” to prompt
3. Different styles created by random feature include photorealism, sci-fi, 3D render, ink wash, meme, glitch, and kawaii
Midjourney has updated its random style feature to work with the latest artificial intelligence image generation model. By adding “–sref random” at the end of your prompt, you can let the AI determine the look and feel of the image entirely. This feature allows for the creation of images in various styles, from anime to film noir and greyscale, providing a way to test the possibilities. For those new to Midjourney, there is a guide available along with a selection of prompts to try out.
To randomize the style of an image, simply add “–sref random” to the end of your prompt. Repeating the prompt with the same parameter will generate a range of images in different styles. This feature is a great way for beginners to explore the AI image generator, showcasing the range of styles available and providing inspiration for future projects. If you find a style you like, you can preserve it by typing “–sref” followed by the URL of the desired image.
Midjourney’s CEO, David Holz, described the new feature as a tool for exploring visual styles and testing new algorithms. The feature was introduced to allow users to experiment with different styles and broaden the capabilities of the AI image generator. Each generated image can be further analyzed with Midjourney Describe to understand how the AI model interprets different styles. Future upgrades may include providing descriptions of the styles generated by the random feature.