17.3.2023Friends, I am excited to share my tests and finding about the newest and most advanced model from Midjourney, the V5 ,released a couple of days ago.

According to the Midjourney technical documentation, the V5 model has several notable advancements over its predecessor.

These include a higher degree of coherency, which allows for more accurate representations of the text prompt in the resulting image.

Additionally, the V5 model excels at interpreting natural language prompts (NLP) and supports advanced features like repeating patterns with the --tile function, which generates interlocking, recurring patterns.

The Midjourney team also claim that V5 is less opinionated, which means that it produces output with fewer influences of embedded styles.

Compared to the V4 model, the V5 produces more realistic images with more detail and fewer errors.

prompt: people playing beach volley

The V5 images are rendered at twice the resolution of V4—1024 X 1024 px compared to 512 X 512 px. Furthermore, the V5 model allows for an unlimited aspect ratio, whereas the V4 only allowed the square aspect ratio at launch.

--ar (aspect ratio)

Another significant advantage of the V5 model is its ability to handle large groups of people and produce more realistic hands, arms, and legs with the right amount of fingers/limbs (most of the time) and fewer random artefacts.

prompt: the pianist hands

Finally, the V5 model offers support for image weights, which lets you assign priority to the image compared to the text part of the prompt.

--iw (weight 0.5 to 2.0 )

I encourage you to try out the Midjourney V5 model and experience its advanced features for yourself.

To get started, join their Discord community and simply add the --v 5 to your prompt.

If you liked this content you can collect this entry and show support!🙌

by @iamsimoleo
Simone Leonelli