Instructions to use Cierpliwy/krea2_lineart_edit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use Cierpliwy/krea2_lineart_edit with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("krea/Krea-2-Raw", dtype=torch.bfloat16, device_map="cuda") pipe.load_lora_weights("Cierpliwy/krea2_lineart_edit") prompt = "-" image = pipe(prompt).images[0] - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- Draw Things
- DiffusionBee
Krea 2 lineart edit

- Prompt
- -
LoRAs for Krea-2-Raw that turn a lineart sketch into a fully rendered image. You provide a lineart (e.g. extracted from a photo with the Realistic Lineart preprocessor) and a text prompt describing the final look — the model paints it while following your lines.
Two variants are included:
| File | Purpose |
|---|---|
krea2_lineart_edit_v1.safetensors |
lineart — follows the sketch most faithfully |
krea2_lineart_coarse_edit_v1.safetensors |
lineart_coarse — looser, more stylized interpretation (used in the example workflow) |
Column 2 vs 3 in the gallery below shows the difference.
Requirements
- ComfyUI
- ComfyUI-Krea2-Ostris-Edit — Krea 2 nodes (
TextEncodeKrea2OstrisEdit,Krea2OstrisEditModelPatch) - comfyui_controlnet_aux — for the Realistic Lineart preprocessor (the lineart format these LoRAs were trained with)
- Krea-2-Raw models:
- UNet:
krea2-turbo-mxfp8-simple.safetensors→ComfyUI/models/unet - CLIP:
qwen3vl_4b_fp8_scaled.safetensors→ComfyUI/models/clip - VAE:
qwen_image_vae.safetensors→ComfyUI/models/vae
- UNet:
Quick start
- Download the LoRA you want (Files & versions tab) and put it in
ComfyUI/models/loras/. - Load the example workflow
lineart_example.jsoninto ComfyUI (Workflow → Open). - In the LoadImage node, drop in your lineart. Tip: run any photo through the Realistic Lineart preprocessor from
comfyui_controlnet_auxfirst — the LoRAs work best with that specific line style. - In the TextEncodeKrea2OstrisEdit node, write a prompt describing the final image (style, colors, setting).
- Queue the prompt.
- Downloads last month
- 138
Model tree for Cierpliwy/krea2_lineart_edit
Base model
krea/Krea-2-Raw