Say Goodbye to Prompt Engineering: Midjourney V6 Guide!

Midjourney has once again levelled the playing field for AI art generators with their latest release of the version V6 and today we will be diving deep into some of the new exciting features of the tool!

🛠️ Useful Links:

📲 Connect with me on other platforms:

P.S: There's a typo in the prompt at 2.47 - the correct prompt for the perfume example is:
a close-up product mid shot of a perfume bottle with a carving of the text "BABCHI" on it, with dark green fluid inside the bottle, lying on the water surface, with flowers, stone on sparkling water, reflections and mirroring, beautiful soft lighting and shadow --ar 16:9 --style raw --v 6.0

Cheers :)

📜 Chapters:
0:00 - Intro
0:31 - MJ V6
1:00 - V6 vs V5 Accuracy Test
2:26 - V6 Text in Image Generation
4:16 - V6 for Storyboarding & Comic
5:43 - V6's Creative Upscale
6:37 - Replicate Any Animation Style

#midjourney #v6 #aiart #aianimation #aifilm

[MorningGloryT]

So much improvement 👍🏻 thank for sharing

[sidhantwaghmare]

Bro please tell that can we also make Doraemon, ninaja hattori image in the same style of 90s, by midjoruney, please tell me, if possible make a video on it😢🎉❤😊

[itsyaboidaniel]

2:51 My man, you put the wrong prompt under the AI image.

[MartinMedvec]

There's an error in the video. At 2:47 the prompt is from the first example.

[paulhiggins]

I think this was inevitable- 'prompt engineering' was always going to be an ephemeral thing in a world where LLM's exist- after all the entire point of a large language model is to be able to understand ordinary language. I'm not sure everyone will be happy about this however- as art generators become more responsive to ordinary language instructions it's going to become harder to argue that any kind of special skill is required to use them.

Ironically while it makes sense for the developers to make their software as easy to use as possible- that very ease of use will render the perceived value of the resulting images as near zero- which may be a problem if you wanted to use AI images in a commercial product, because the message you will be sending by using AI art is that your product was made using 'low effort' art, which may not be the message you want potential customers to receive. In this sense the interests of the developers-to maximise their user base- and the interests of some of their users- to make money from creating AI Art- are in conflict.

To be seen as having value AI Art needs some kind of 'mystique' - some idea- at least -that creating it is marginally more difficult than falling off a log. But since ease of use is a major objective of it's developers this seems unlikely. The prompt engineer meme was in part an attempt to create such a mystique and for a while looked like it might work- but as the natural language abilites of generative software of all kinds increases it's becoming clear that there will be no new technocratic class of 'Prompt Engineers'- using AI's will not be the sole province of skilled elites with the specialised arcane knowledge required to optimise them. Indeed the inevitable end game here is rather that AI's themselves will be used as interlocuters to refine and optimise the prompts fed into other AI's- wth the role of the human operators being more akin to the Art Director rather than the Artist.