Instructions to use CtrlAltArt/Flux_Coloring_Pages_lora with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use CtrlAltArt/Flux_Coloring_Pages_lora with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("black-forest-labs/FLUX.1-dev", dtype=torch.bfloat16, device_map="cuda") pipe.load_lora_weights("CtrlAltArt/Flux_Coloring_Pages_lora") prompt = "-" image = pipe(prompt).images[0] - Inference
- Notebooks
- Google Colab
- Kaggle
- Local Apps
- Draw Things
- DiffusionBee
prompt
#1
by lukasor - opened
what prompt do you use to make such a clear image? may i know the ksampler, scheduler?
I used simple prompts for the sample images, the ship was "coloring book of a pirate ship on the high seas", octopus was "coloring book close-up of a kraken's face" and so on, The workflow I used can be seen by dragging any of the images in to a comfy tab, kind of outdated now with the new subgraph update but its a basic flux workflow and then the image is upscaled using the Comfyui_UltimateSDUpscale node. The Ultimate SD Upscale node is what makes it so sharp. Sampler name used was euler, scheduler simple, 40 steps
CtrlAltArt changed discussion status to closed