Text-to-Image
Diffusers
Safetensors
English
Text-to-Image
ControlNet
Diffusers
Flux.1-dev
image-generation
Stable Diffusion
Instructions to use Shakker-Labs/FLUX.1-dev-ControlNet-Depth with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use Shakker-Labs/FLUX.1-dev-ControlNet-Depth with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("Shakker-Labs/FLUX.1-dev-ControlNet-Depth", dtype=torch.bfloat16, device_map="cuda") prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k" image = pipe(prompt).images[0] - Notebooks
- Google Colab
- Kaggle
- Local Apps
- Draw Things
- DiffusionBee
Bad result if I make controlnet_conditioning_scale value bigger
#2
by chuckma - opened
If controlnet_conditioning_scale is 0.7, prompt like "A man in Times Square, New York", image quality will be quite bad. How to avoid this while keeping controlnet_conditioning_scale at 0.7?
We also noticed that a large scale can lead to very noisy result. Try to decrease the scale, 0.5 should also be fine for human generation.