Exploring DNSR, Midjourney, and Leonard AI for my image generation

I have always wanted to use midjourney but it usually sounds and looks complicated. I procrastinated as with everything else with me until I recently joined Discord. I dumped it when the whole scene looked rather complicated.

Image generated with Leonard AI
Today, I decided to stop being my enemy and do something. First off, I went to Shadows' blog to stalk. Hoping to find a few guides on how she utilizes midjourney. And then I found that she worked on the daily shadows pub prompt for the day using midjourney. And I decided to check my mail if I have a recent one from them. There I had one. And it was the exact word, resilience. Prompt A Day -- Wednesday, June 28th - Resilience

While reading her blogs, I found a few things that did not make any sense to me until I googled how to use midjourney to generate images. The blog did a good job of explaining and at that point, everything Shadows said in her resilience post began to make sense.

With everything I found, I started on my midjourney. The first time I wanted to use midjourney I was confused about the different newbie channels available for generation. Today everything made sense when I did extensive reading. So I tried to generate by direct DM to the bot and I was told there were lots of demands and I should either subscribe or come back tomorrow.

That was annoying because I really wanted to learn to use the AI today. I decided to use one of the available newbie channels in case the error message was for just DMs and I found out it doesn't matter where I stood I won't be able to use midjourney today except if I pay.

Surely there has to be a way around this, I thought, and that's how I decided to stalk the AI community on Hive to see if anyone has done anything different from Shadows, probably someone who is using the free version of Midjourney. It was in the course of this research that I found this guy, Brando, who introduced me to Leonardo.ai and now this blog post is a success.

When he listed AI apps that we can utilize to do stuff and he included image generation, I was indeed glad.

I signed up with my mail, created a username and I was ready to generate.

I have Shadow to thank for introducing the DNSR to me as part of the midjourney generation. She uses it for her prompts and I did too. I had the result saved so it was easy to pick it up and stuff it into Leonardo.ai and get results.

What DNSR did was asked what I wanted to do in a step-by-step format and I provided answers, he told me to pick whatever he gave me and give it to midjourney and I'll have my images. What DNSR did not know was that I had no subscription to midjourney and that at the end of the day, I'll be using a different AI tool to get what I wanted.

DNSR helped because I did not have to use my limited 150 credits to ask Leonardo.ai for prompt ideas. It was free on DNSR so that step was excused.

I resisted the urge to try different things with the results I got for my resilience image because I had limited credits. So I am presenting what came out without a tweak. And I'll tell you my favorite at the end of this blog. You can pick a favorite and share it with me.

Let's do a run down on the steps. First I use DNSR to generate my promote for midjourney using the command, /design. And he asked the questions shown on the screenshots.

I chose photorealistic for the question

I chose vibrant and energetic for this one.

I chose determined and focused for the third question.

I chose leaning forward with anticipation for the 4th question

Knowing that I often see Pixaby return images with people in suit standing close to skyscrapers if you search for successful people, I decided to choose Urban Cityscapes with skyscrapers for this question.

After generating DNSR gave me further instructions to pick the prompt and generate images using /imagine on midjourney which you know failed as a result of the queues and no subscription.

Let's see what Leonardo.ai gave me after signing up and adding my prompt.

I did not change any of the fine tuned model since it was first time but I did choose Leonard style since it rang a bell with the name of the AI.

The AI used 8 credits out of 150 credits and gave me these four results

I did not like the first three results because of the mark on his face. I preferred the last one not only because it gave me what I wanted which was, "I have the word resilience and I want a character who hustles his way to the top against all odds" but because it had a great view of the skyscraper I had in mind all along.

Which of the images is your favorite? If you are struggling with midjourney you can at least use Leonardo.ai to generate the images of your choice. And of course, if you have a lot of credits you can tweak them as you play around with the AI.