General Image Generation

docker pull cargoshipsh/image-generation-sd

Automatically generates any kind of image. This Stable Diffusion model was created by the researchers and engineers from CompVis, Stability AI, runway and LAION. The StableDiffusionPipeline is capable of generating photo-realistic images given any text input using Stable Diffusion. On a 2060s GPU it takes only a few seconds to generate an image.

Demo

Coming Soon

License

The model is licensed under the Apache 2.0 License and the code for the API wrapper is licensed under MIT License.

System Requirements

Minimum: 8GB RAM, Nvidia GPU with 6GB VRAM, 10GB disk space
Recommended: 16GB RAM, Nvidia GPU with 16GB VRAM ,100GB disk space

Usage

Input [POST]

The input expects a prompt and an optional neg_prompt as strings. The prompt should be what you want to create. The neg_prompt or negative prompt should contain things you don't want to appear in the picture.

{
  "prompt": "a black cat drinking coffee, photorealistic, realistic",
  "neg_prompt": "cartoon, comic, bright, artifacts, bad art, poor quality"
}

Output

The output is a 512x512 pixel png-image as bytearray

You need to set an API Key via the environment variable API_KEY to run the image and set the X-API-KEY header in your request with the same KEY.

Example

Before you run the container, make sure you have Docker and the Nvidia Docker toolkit installed. Visit NVIDIA Docker for more detailed information.

Then run the following command:

docker run -p 80:80 --env API_KEY=CHANGE_ME cargoshipsh/image-generation-sd

In a new terminal window, run the following command to call the API

curl -X POST -H "X-Api-Key: CHANGE_ME" -H "content-type: application/json" -d '{"prompt": "a black cat drinking coffee, photorealistic, realistic", neg_prompt": "cartoon, comic, bright, artifacts, bad art, poor quality"}' localhost:8000 > cat.png

The image is saved in your current directory.