I have a project I have been working on for about 80 hours or so now. I am trying to create a graphic novel. I have been working for some time with ChatGPT and OpenAI in an attempt to render consistent images but my attempts all end in failure. I seem to have a “tug of war” going on. If I forget architecture I can render images with a >90% consistency art style, if I forget art style I can render images with a >90% consistency architecture with a minimum amount of drift to specific identifiable artifacts.
But I can’t seem to find the correct verbage to achieve both affects simultaneously. I have in my efforts created nearly an entire Production Bible with hard rules, laws, a hierarchy and it seems I’ve just hit a dead end.
Have I reached the limits of what OpenAI can offer in its current iteration? Or just the limits of myself?
Welcome to the dev Community, @Skeletor1
Thanks for reporting this. There was a known issue affecting image generation around the time you experienced this, which may explain the behavior you saw.
Is this still happening for you, or has it started working again? Also, are you using the Free plan or a paid plan (Plus/Pro)? That will help provide a bit more context.
If the issue is still happening, please reach out to support@openai.com with the details so the team can take a closer look.
~ Smith
Hey there. I am using the $20 a month plan. And consistant renderings (especially across multiple threads as I keep hitting RAM constraints) is an issue
Hey there. I am using the $20 a month plan. And consistant renderings (especially across multiple threads as I keep hitting RAM constraints) is an issue
Thanks for the additional information.
I understand you're still experiencing issues with maintaining consistent image generation throughout your project.
The best next step would be to open a support ticket so our team can take a closer look at what you're experiencing.
Thanks again for following up, and I hope this gets resolved soon.