https://github.com/AUTOMATIC1111/stable-diffusion-webui/wiki/API
If you have any backends that need to be adapted, please feel free to leave an issue, and I will work on the adaptation.
This project plans to implement i18n. (woking on it)
checkpoint by/模型 https://huggingface.co/nyanko7/flux-dev-anime-cg
- A Simple Gradio GUI, You can change your model here(Running on FastAPI port + 1)
- Multi-backend load balancing
- Automatic locking for backends that do not support concurrency
- Built-in caption feature(https://github.com/toriato/stable-diffusion-webui-wd14-tagger), currently supporting natural language tagging (CUDA/8G VRAM) and wd tagger
- https://github.com/AUTOMATIC1111/stable-diffusion-webui
- https://civitai.com/
- https://fal.ai/models/fal-ai/flux/schnell
- https://replicate.com/black-forest-labs/flux-schnell
- https://www.liblib.art/
- https://tusiart.com/
- https://www.seaart.ai/
- https://www.yunjie.art/
- https://github.com/comfyanonymous/ComfyUI
- https://novelai.net/image
- https://github.com/novicezk/midjourney-proxy
Python 3.10
git clone https://github.com/DiaoDaiaChan/Stable-Diffusion-DrawBridgeAPI
cd Stable-Diffusion-DrawBridgeAPI
python -m venv venv
.\venv\Scripts\python -m pip install -r .\requirements.txt
Copy config_example.yaml to config.yaml.
See Detailed Instructions
.\venv\Scripts\python.exe -m DrawBridgeAPI.api_server --port=8000 --host=127.0.0.1 --conf .\DrawBridgeAPI\config.yaml
Visit http://localhost:8000/docs# for help.
The API currently has no authentication; do not expose this API to the public, or it may be abused.
Testing with CURL
curl -X POST -H "Content-Type: application/json" -d '{"prompt": "reimu", "width": 512, "height": 768}' http://localhost:8000/sdapi/v1/txt2img
..\venv\Scripts\python api_server.py --port=8000 --host=127.0.0.1
Start the built-in tagging server.
Set server_settings - build_in_tagger to true in the config.yaml file to start it, and install the dependencies.
If the working directory is Stable-Diffusion-DrawBridgeAPI\DrawBridgeAPI:
..\venv\Scripts\python -m pip install -r utils/tagger-requirements.txt
- A simple frontend for easy drawing, tagging, and other operations.
Updated to use LLM for natural language image tagging, adapted from https://github.com/StartHua/Comfyui_CXH_joy_caption.
Note: Requires GPU and 8G VRAM.
build_in_tagger:
true
llm_caption: # Use LLM for natural language tagging
enable:
true
clip:
google/siglip-so400m-patch14-384
llm:
unsloth/Meta-Llama-3.1-8B-bnb-4bit
image_adapter: # https://huggingface.co/spaces/fancyfeast/joy-caption-pre-alpha/tree/main/wpkklhc6
image_adapter.pt
Dependencies are in llm_caption_requirements.txt