If you're only after the image, you can pretty much use any tool that creates images, including the Genspark one I covered in this post. Here's what I just got from four models for a quick "book cover featuring a tiger looking up at the moon" test:
You're a MJ wizard, but I'm most interested in the hot takes. I get a ton of value from when you quickly review something and tell us what sucks about it and what's promising, and of course we have this firehose of information problem that can be whittled down a bit with a hot take.
Also: they can be pretty funny. I think they'll continue to be a hit, and folks can unsubscribe from that aspect anyway.
Thanks for the feedback. That's the plan: test drive promising new tools or features to figure out how you can actually use them in your daily life.
As for the video guides, they're definitely going to be about more than Midjourney. This one just happened to be about MJ, but the idea is that I'll look at what people are asking about and tackle those questions through "How to" guides.
Sounds like a plan! I think I've mentioned it before, but even the MJ-specific guides have been helpful, just to see the thought process behind prompting and to better understand how models in general work. Looking forward so exploring some new tools this year.
I tried using the --p value you used it worked for me, looks like it is public.
Quick tip: you drag/dropped images to the prompt bar to apply them. I find it easier to apply a cref from a previous prompt by clicking the thumbnail in the right, just like re-using the text prompt.
Awesome, thanks for confirming that the moodboard code works!
As for your tip, are you referring to when I used one of the "--chaos 100" images as a style reference? If so, there was no --sref code to reuse, since the image wasn't discovered through --sref random.
Having said that, one could also click the "Use => Style" button in the bottom-right "Actions" panel next to the image to use it as a style reference. But I wanted to visually showcase the three options of reusing an image (image prompt, character reference, style reference), hence the drag-and-drop shenanigans.
Or are you referring to something else? Maybe there's something I've overlooked!
Hi Daniel. Sounds like fun - which is my main motivation. I must own up to still being amazed at the rate of change in these tools over the last year or so (I don’t think that it’s just my age!) All the best, John.
I agree, John. I stay up to date regularly because of my work for this newsletter, but even I'm overwhelmed by the pace of innovation, news, etc. It's a LOT!
Do let me know if you end up trying Genspark - would be curious if you had some fun results from it.
Hi! Do you know of a way to use AI to create a book cover?---even if it costs some money? I just need some pictures. I would do the words myself.
Hey Noah,
If you're only after the image, you can pretty much use any tool that creates images, including the Genspark one I covered in this post. Here's what I just got from four models for a quick "book cover featuring a tiger looking up at the moon" test:
https://i.imgur.com/V7LPkWI.png
(I set the aspect ratio to 2:3 in this case)
If that doesn't cut it, you can use many of these image models directly. I looked at some of them in this now-somewhat-outdated post: https://www.whytryai.com/p/text-to-image-ai-models
Finally, you can even try to get them to take a stab at the text. I tested "spelling" image models here: https://www.whytryai.com/p/ai-image-model-spelling-text
So basically, just use your preferred image model, request a book cover, describe the setting/scene, and set the aspect ratio to the desired format.
Hope that helps - good luck!
Thank you so much!! I'll look into those! That's so helpful. Thank you!
You got it, happy to help!
You're a MJ wizard, but I'm most interested in the hot takes. I get a ton of value from when you quickly review something and tell us what sucks about it and what's promising, and of course we have this firehose of information problem that can be whittled down a bit with a hot take.
Also: they can be pretty funny. I think they'll continue to be a hit, and folks can unsubscribe from that aspect anyway.
Thanks for the feedback. That's the plan: test drive promising new tools or features to figure out how you can actually use them in your daily life.
As for the video guides, they're definitely going to be about more than Midjourney. This one just happened to be about MJ, but the idea is that I'll look at what people are asking about and tackle those questions through "How to" guides.
Excited to be trying something new this year!
Sounds like a plan! I think I've mentioned it before, but even the MJ-specific guides have been helpful, just to see the thought process behind prompting and to better understand how models in general work. Looking forward so exploring some new tools this year.
Great post, thanks!
I tried using the --p value you used it worked for me, looks like it is public.
Quick tip: you drag/dropped images to the prompt bar to apply them. I find it easier to apply a cref from a previous prompt by clicking the thumbnail in the right, just like re-using the text prompt.
Awesome, thanks for confirming that the moodboard code works!
As for your tip, are you referring to when I used one of the "--chaos 100" images as a style reference? If so, there was no --sref code to reuse, since the image wasn't discovered through --sref random.
Having said that, one could also click the "Use => Style" button in the bottom-right "Actions" panel next to the image to use it as a style reference. But I wanted to visually showcase the three options of reusing an image (image prompt, character reference, style reference), hence the drag-and-drop shenanigans.
Or are you referring to something else? Maybe there's something I've overlooked!
Hi Daniel. Sounds like fun - which is my main motivation. I must own up to still being amazed at the rate of change in these tools over the last year or so (I don’t think that it’s just my age!) All the best, John.
I agree, John. I stay up to date regularly because of my work for this newsletter, but even I'm overwhelmed by the pace of innovation, news, etc. It's a LOT!
Do let me know if you end up trying Genspark - would be curious if you had some fun results from it.