秒画
Miaohua is a free AI painting and image generation platform from SenseTime, supporting text-to-image, image-to-image, ControlNet precision control and custom LoRA training, with output up to 6K resolution.
Human verified · · Submit a correction
Ideal for designers, marketers and ACG creators who want free, high-resolution image generation; less suitable for advanced users needing private deployment and deeply customized pipelines.
Verified facts
What is 秒画
Miaohua is a free AI painting and image generation platform from SenseTime, supporting text-to-image, image-to-image, ControlNet precision control and custom LoRA training, with output up to 6K resolution.
Key features of 秒画
- Generate high-definition images up to 6K from text prompts
- Upload an image and restyle it into similar works in other styles
- Use ControlNet for pose control, line-art coloring and depth detection
- Train a custom LoRA model from local images for personalized output
Good for
- SenseTime's in-house painting model with up to 6K output
- Complete feature set: text-to-image, image-to-image, ControlNet and LoRA training
- Free to use after registration
Watch out
- Output quality depends heavily on prompt quality, so beginners face a learning curve
- Generation speed may slow during peak platform load
- Custom LoRA training requires users to prepare compliant image material themselves
How to use 秒画
- Open miaohua.sensetime.com and click login
- Register or sign in with your email
- Enter the creation interface and click start
- Choose quantity, resolution, base model and LoRA style on the left
- Type your prompt on the right and click generate
Who 秒画 is for
Difficulty: Beginner
- Generate high-definition images up to 6K from text prompts
- Upload an image and restyle it into similar works in other styles
- Use ControlNet for pose control, line-art coloring and depth detection
- Train a custom LoRA model from local images for personalized output
FAQ
Is Miaohua free?
Yes. Miaohua is a free AI painting platform from SenseTime; register an account and you can start creating.
What is the maximum output resolution?
According to the official introduction, text-to-image generation can reach up to 6K high definition.
Can I train my own model?
Yes. You can upload local images and train a custom LoRA model based on SenseTime's in-house models or open-source models.
Sources and verification
Sources: miaohua.sensetime.com (opens in a new tab)
Verified: · Submit a correction →
Alternatives to 秒画
All in this category
Bing Image Creator
Great for casual users and content creators who want a zero-learning-curve way to generate images from text; less suitable for professional designers who need fine-grained style control and production-grade output.
Zyro AI Background Remover
Best for small sellers, entrepreneurs, and Zyro users handling straightforward images one at a time; not suited to batch work, API integration, or precision edge repair.
Zyro AI Image Upscaler
Good for Zyro site builders and casual users who want a free, fast way to enlarge images; not suited for heavy users needing extreme magnification, batch processing or professional controls.
Media.io AI Image Upscaler
Good for existing Media.io users who also need image upscaling, and for casual users wanting a quick online enlarger; not suited for professional scenarios needing extreme magnification or offline batch processing.