I’m trying to generate an image and I’ve tried a few generators. Out of the 10 or so I’ve tried, only Leap will actually place items where I want them.
Eg: vaporwave wireframe mountains with a purple sun in the upper right corner of the image
Most of the time every generator will put the sun in the middle of the image.
Is it a matter of being bad at asking for what I want? Is it just that most of them won’t follow simple commands about where things should go?
Any advice would be greatly appreciated!
Right now the best model is Flux: https://huggingface.co/spaces/black-forest-labs/FLUX.1-schnell
You can also get their pro version if you are ok with paying: https://blackforestlabs.ai/
Come back in 6 months and the answer will be different. Things are moving very rapidly.
Remember that tons of website are basically running the same code (usually stable diffusion) under the hood and just provide you their own UI. Then you have Dall-E (In Bing image generator) and Midjourney which have their own code under the hood.
Also, beside a bit of luck, you can define how much you want the model to follow your prompt and other similar stuff, let alone the whole prompting part.
☞ Region-specific prompts
Do you know you can specify different prompts for different regions of the image?
For example, you can put the moon at the top left: https://stable-diffusion-art.com/prompt-guide/#Region-specific_prompts
Thank you!!
Have you tried Flux?
I’ve heard it’s one of the best, currently.
In answer to your question: Mine.
Mine’s the best.