From the course: Midjourney: Tips and Techniques for Creating Images

Working with Midjourney v6 - Midjourney Tutorial

From the course: Midjourney: Tips and Techniques for Creating Images

Start my 1-month free trial Buy for my team

Working with Midjourney v6

“

- [Instructor] At the time of this recording, Midjourney is currently on V6 Alpha. You can enable this under settings or alternatively, add the parameter --V6, after your text prompt. Some of the changes include Upscales are two times faster, so you're going to be saving time on GPU minutes. You're going to see improved aesthetics, coherence, prompted adherence, and image quality. However, I recommend that you experiment and check out the results yourself. We're also going to take a look at improved text rendering, which you can now define by putting the text that you want to see in your image in brackets, not to mention a style reference. Now, since this is an Alpha, Midjourney has warned that this could change quite a bit. If it is the default model, you'll see it available right away. But if it's not, head into Discord and type in /settings. And from the settings list, you can enable Midjourney Model V6 or the Niji Model V6 to be the default. Now, when I enter an imagine prompt, such as the lunar new year in 2025 captured from a boat at night using a 30 millimeter lens, following that with a comma, follow that with the aspect ratio of 2:1, once I press return, you'll see that V6 gets appended to the end. And you can do that of course, manually. And here's what Midjourney produced. Let's actually go to Upscale and see how quickly that is produced. Another way to check out the quality of V6 is to rerun images produced with earlier models. I'm currently here on midjourney.com. I'm searching all of the images that I produced. I found this cat image produced with version V5. What I can do is go to the options menu and copy the prompt that I used, head to Discord. Let's use imagine followed by the cat wearing sunglasses, and I'll remove the V5 appendix and then just let V6 come up naturally. Here are the results of V6. I definitely see a few of these images representing a sad cat. And if I compare this to the original image, which I'm not sure he was looking sad, more contemplative in this shot, I'll compare the quality of the bananas and the lemon next to the ones in this new version. So have some fun experimenting with V6. And in the next few movies, we're going to cover the new text feature, not to mention the style reference parameter.

Contents