Coloring Book Flux
Coloring Book Flux is a Flux LoRA trained on a 100 image synthetic dataset that I personally generated, the images in the dataset were mostly human, vehicles and animal illustrations. All of these images were captioned using Joy Caption Batch.
FOR THE BEST RESULTS USE DEIS SAMPLER!
This has been a pet project of mine for a while and it took lots of trial and error to get it to where it is at this stage. I iterated through many failed attempts until finally I began getting consistent results by limiting repeats, increasing epochs and lowering DIM/ALPHA settings in Kohya.
One neat surprise with this one is that you can prompt for colored images in this style as well. When prompting for coloring book pages it might help to explicitly mention a white background.
This can be a very useful resource for coloring books, posters, greeting cards, print on demand, stock imagery, clip art and so much more.
I hope you have as much fun with this as I did creating it!
You have to work a little bit to get desired results and sometimes there is bleeding/blending of subjects but overall the style is present and the results can be really good. This LoRA takes a couple of tries adjusting your prompt and adding tokens to match the style.
If you enjoy this resource please send BUZZ ⚡️so I can keep experimenting and sharing. Please share your images in the gallery after downloading them, by sharing them in the gallery the system gives creators buzz to help with generating and training.
Trigger keyword(s): 'c0l0ringb00k' 'coloring book page'
Recommended strengths: 0.7 - 0.9
Description
0.0001 Learning Rate, 2 Repeats, 10 Epochs, 2000 steps, DIM 16, ALPHA 32
FAQ
Comments (6)
This is awesome, good job :) I noticed you had a lot of trying to get this LoRA to work as best as possible.. was there anything you learned particularly that worked best/better? what settings and whatnot?
Thanks man, for these latest LoRAs I shared setting learning rate set low and adjusting DIM and Alpha helped with simple styles. I think everyone defaults to Euler when generating and it appears to me that DEIS is better right now. Another thing is not going crazy on repeats and relying more on epochs instead.
has anybody tried this with an image-to-image or canny controlnet? I'm trying to transform a colored image to a coloring book page
I'm trying the same thing, but it seems to have no effect.
Try "line Art", Pixaroma on YouTube shows it.
Very nice lora. Thank you for this. For me its the best lora for coloring books.
Details
Files
c0l0ringb00k_Flux_v1_renderartist.safetensors
Mirrors
c0l0ringb00k_Flux_v1_renderartist.safetensors
c0l0ringb00k_Flux_v1_renderartist.safetensors
c0l0ringb00k_Flux_v1_renderartist.safetensors
c0l0ringb00k_Flux_v1_renderartist.safetensors
c0l0ringb00k_Flux_v1_renderartist.safetensors
c0l0ringb00k_Flux_v1_renderartist.safetensors
c0l0ringb00k_Flux_v1_renderartist.safetensors
Available On (2 platforms)
Same model published on other platforms. May have additional downloads or version variants.











