---
tags:
- text-to-image
- lora
- diffusers
- template:diffusion-lora
widget:
- text: '-'
  output:
    url: images/Coloring_Book_00001_.png
- text: '-'
  output:
    url: images/Coloring_Book_00002_.png
- text: '-'
  output:
    url: images/Coloring_Book_00004_.png
- text: '-'
  output:
    url: images/Coloring_Book_00011_.png
- text: '-'
  output:
    url: images/Coloring_Book_00013_.png
- text: '-'
  output:
    url: images/Coloring_Book_00026_.png
- text: '-'
  output:
    url: images/Coloring_Book_00028_.png
- text: '-'
  output:
    url: images/Coloring_Book_00032_.png
- text: '-'
  output:
    url: images/Coloring_Book_00036_.png
- text: '-'
  output:
    url: images/Coloring_Book_00045_.png
- text: '-'
  output:
    url: images/Coloring_Book_00046_.png
- text: '-'
  output:
    url: images/Coloring_Book_00049_.png
- text: '-'
  output:
    url: images/Coloring_Book_00055_.png
- text: '-'
  output:
    url: images/Coloring_Book_00058_.png
- text: '-'
  output:
    url: images/Coloring_Book_00079_.png
- text: '-'
  output:
    url: images/Coloring_Book_00084_.png
- text: '-'
  output:
    url: images/Coloring_Book_00098_.png
- text: '-'
  output:
    url: images/Coloring_Book_00101_.png
base_model: HiDream-ai/HiDream-I1-Full
instance_prompt: c0l0ringb00k, coloring book
license: apache-2.0
---
# Coloring Book HiDream

<Gallery />

## Model description 

This HiDream LoRA is Lycoris based and produces great line art styles similar to coloring books. I found the results to be much stronger than my Coloring Book Flux LoRA. Hope this helps exemplify the quality that can be achieved with this awesome model. This is a huge win for open source as the HiDream base models are released under the MIT license.

I recommend using LCM sampler with the simple scheduler, for some reason using other samplers resulted in hallucinations that affected quality when LoRAs are utilized. Some of the images in the gallery will have prompt examples.

Trigger words: c0l0ringb00k, coloring book

Recommended Sampler: LCM

Recommended Scheduler: SIMPLE

This model was trained to 2000 steps, 2 repeats with a learning rate of 4e-4 trained with Simple Tuner using the main branch. The dataset was around 90 synthetic images in total. All of the images used were 1:1 aspect ratio at 1024x1024 to fit into VRAM.

Training took around 3 hours using an RTX 4090 with 24GB VRAM, training times are on par with Flux LoRA training. Captioning was done using Joy Caption Batch with modified instructions and a token limit of 128 tokens (more than that gets truncated during training).

The resulting LoRA can produce some really great coloring book styles with either simple designs or more intricate designs based on prompts. I&#39;m not here to troubleshoot installation issues or field endless questions, each environment is completely different.

I trained the model with Full and ran inference in ComfyUI using the Dev model, it is said that this is the best strategy to get high quality outputs.

Testing and training takes a lot of time and personal resources. If you can afford it please contribute to my KoFi (https:&#x2F;&#x2F;ko-fi.com&#x2F;renderartist) – Contributing will allow me more flexibility to train in the cloud and continue experimenting and sharing better.

renderartist.com

## Trigger words

You should use `c0l0ringb00k` to trigger the image generation.

You should use `coloring book` to trigger the image generation.


## Download model

Weights for this model are available in Safetensors format.

[Download](/renderartist/coloringbookhidream/tree/main) them in the Files & versions tab.