At this point I'm not sure I can even justify the cost considering the direction. V 3 was perhaps the most interesting point, the artefacts and style built in were so unique. Since then Midjourney has slowly exchanged unique flavour in favour of a so-called bland realism.
Not seeing large degree of improvement that many were expecting between #midjourney v5 and v6. In terms of an understanding of concepts within your text prompt it is perhaps slightly better. It seems to be trained ever more towards dull aesthetics that you have to work against
#aiart As usual the left image used with /describe to create the second image.
In this case the left image was a blend of 3 different #midjourney images I made. Prompt in alt for right image (interesting terms here, speed painting and Necronomicon illustration):
https://t.co/kINksgmtGz
Keep in mind it's a demo and you can't use creations for commercial use. I just like it as an extra way to make my Midjourney images somehow unique. I already prompt in ways that are often not replicable due to remixing and weights.
#midjourney#AIart
Meta new tool called Segment Anything can make collages from #midjourney images(not for commercial use)
Below a 'mockup' image I made. Segment Anything can seperate elements. You can then use those as elements in Canva to create a collage. See second example collage below👇👇👇
This is several random pieces from Midjourney images. I segmented using the Segment Anything tool. Then opened Canva and used those images to create this. This is not a 'good' piece of art just a proof of concept. #midjourneyv5
An extra way to make Midjourney images unique.