SD3.5M-Controlnet-Canny Open-Source Image Generation Model - Precision Image Control with Canny Edge Detection Maps

SD3.5M Controlnet Canny

Developed by tensorart

ControlNet Canny model based on Stable Diffusion 3.5 Medium for controlled image generation via Canny edge detection maps

Image Generation EnglishOpen Source License:Other #Edge-controlled image generation #Rapid iterative optimization #Chinese prompt support

Downloads 44

Release Time : 1/6/2025

Model Overview

This model combines the image generation capability of Stable Diffusion 3.5 Medium with the edge control capability of ControlNet's Canny, enabling generation of compliant images based on input edge maps

Model Features

Edge-controlled generation

Precisely control the composition and shape of generated images using Canny edge detection maps

High-quality generation

Generates high-quality images based on the Stable Diffusion 3.5 Medium model

Fast inference

Supports Turbo version, capable of producing high-quality images in under 8 steps

Model Capabilities

Text-to-image generation

Edge-guided image generation

High-quality art creation

Rapid image generation

Use Cases

Art creation

Concept art design

Artists can use edge maps to quickly generate concept art that meets compositional requirements

Artworks with precisely controlled composition

Product design

Product prototype visualization

Designers can quickly generate product prototype visualizations using simple edge maps

Rapidly iterated design solutions

🚀 TensorArt Stable Diffusion 3.5 Medium ControlNet Canny

A text - to - image model that combines Stable Diffusion 3.5 Medium with ControlNet Canny for enhanced image generation.

🚀 Quick Start

With SD3.5M

import torch
from diffusers import StableDiffusion3ControlNetPipeline
from diffusers import SD3ControlNetModel
from diffusers.utils import load_image

controlnet = SD3ControlNetModel.from_pretrained("tensorart/SD3.5M-Controlnet-Canny")
pipe = StableDiffusion3ControlNetPipeline.from_pretrained(
    "stabilityai/stable-diffusion-3.5-medium",
    controlnet=controlnet
)
pipe.to("cuda", torch.float16)

control_image = load_image("https://huggingface.co/tensorart/SD3.5M-Controlnet-Canny/resolve/main/asset/canny.png")
prompt = "An eye"
negative_prompt = "low quality, worst quality, deformed, distorted, disfigured, motion smear, motion artifacts, fused fingers, bad anatomy, weird hand, ugly, monochrome"

image = pipe(
    prompt, 
    num_inference_steps=30,
    negative_prompt=negative_prompt, 
    control_image=control_image, 
    guidance_scale=4.5,
    controlnet_conditioning_scale=0.8
).images[0]
image.save('image.jpg')

With TensorArt's SD3.5M Turbo

import torch
from diffusers import StableDiffusion3ControlNetPipeline
from diffusers import SD3ControlNetModel
from diffusers.utils import load_image

controlnet = SD3ControlNetModel.from_pretrained("tensorart/SD3.5M-Controlnet-Canny")
pipe = StableDiffusion3ControlNetPipeline.from_pretrained(
    "tensorart/stable-diffusion-3.5-medium-turbo",
    controlnet=controlnet
)
pipe.to("cuda", torch.float16)

control_image = load_image("https://huggingface.co/tensorart/SD3.5M-Controlnet-Canny/resolve/main/asset/canny.png")
prompt = "An eye"
negative_prompt = "low quality, worst quality, deformed, distorted, disfigured, motion smear, motion artifacts, fused fingers, bad anatomy, weird hand, ugly, monochrome"

image = pipe(
    prompt, 
    num_inference_steps=8,
    negative_prompt=negative_prompt, 
    control_image=control_image, 
    guidance_scale=1.5,
    controlnet_conditioning_scale=0.8
).images[0]
image.save('image.jpg')

📚 Documentation

Contact

Website: https://tensor.art https://tusiart.com
Developed by: TensorArt

📄 License

License: other
License Name: stabilityai-ai-community
License Link: LICENSE.md

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご