Stable Diffusion WebUI Has a Free API — Generate AI Images Locally

AUTOMATIC1111's Stable Diffusion WebUI is the most popular interface for running AI image generation locally. It has a built-in REST API that lets you generate, edit, and upscale images programmatically.

Free, open source, runs on your GPU. No API costs, no content restrictions.

Why Use the SD WebUI API?

• Free generation — no per-image costs
• Full control — any model, any LoRA, any settings
• No content policy — generate whatever you want (legally)
• Batch generation — produce hundreds of images programmatically
• All features — txt2img, img2img, inpainting, upscaling, ControlNet

Quick Setup

1. Install

git clone https://github.com/AUTOMATIC1111/stable-diffusion-webui
cd stable-diffusion-webui

# Start with API enabled
./webui.sh --api
# API runs on http://localhost:7860

2. Text to Image

curl -s -X POST http://localhost:7860/sdapi/v1/txt2img \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "a futuristic data center with glowing servers, cyberpunk style, 4k",
    "negative_prompt": "blurry, low quality",
    "steps": 30,
    "width": 768,
    "height": 512,
    "cfg_scale": 7,
    "sampler_name": "DPM++ 2M Karras"
  }' | python3 -c "
import json, sys, base64
data = json.load(sys.stdin)
img = base64.b64decode(data['images'][0])
with open('output.png', 'wb') as f: f.write(img)
print('Saved output.png')
"

3. Image to Image

# Convert image to base64 first
IMG_B64=$(base64 -i input.png)

curl -s -X POST http://localhost:7860/sdapi/v1/img2img \
  -H "Content-Type: application/json" \
  -d "{
    \"init_images\": [\"$IMG_B64\"],
    \"prompt\": \"oil painting style, masterpiece\",
    \"denoising_strength\": 0.5,
    \"steps\": 30
  }" | python3 -c "
import json, sys, base64
data = json.load(sys.stdin)
img = base64.b64decode(data['images'][0])
with open('img2img_output.png', 'wb') as f: f.write(img)
print('Saved img2img_output.png')
"

4. Get Available Models

# List models
curl -s http://localhost:7860/sdapi/v1/sd-models | jq '.[].title'

# List samplers
curl -s http://localhost:7860/sdapi/v1/samplers | jq '.[].name'

# List LoRAs
curl -s http://localhost:7860/sdapi/v1/loras | jq '.[].name'

# Current options
curl -s http://localhost:7860/sdapi/v1/options | jq '{model: .sd_model_checkpoint, vae: .sd_vae}'

5. Switch Model

curl -s -X POST http://localhost:7860/sdapi/v1/options \
  -H "Content-Type: application/json" \
  -d '{"sd_model_checkpoint": "v1-5-pruned-emaonly.safetensors"}'

Python Example

import requests
import base64
from pathlib import Path

SD_URL = "http://localhost:7860"

# Generate image
response = requests.post(f"{SD_URL}/sdapi/v1/txt2img", json={
    "prompt": "a robot analyzing data on multiple screens, digital art",
    "negative_prompt": "blurry, ugly, deformed",
    "steps": 30,
    "width": 768,
    "height": 512,
    "cfg_scale": 7,
    "batch_size": 4  # Generate 4 images at once
}).json()

for i, img_b64 in enumerate(response["images"]):
    img_data = base64.b64decode(img_b64)
    Path(f"output_{i}.png").write_bytes(img_data)
    print(f"Saved output_{i}.png")

# Get generation info
info = requests.post(f"{SD_URL}/sdapi/v1/png-info",
    json={"image": response["images"][0]}).json()
print(f"Generation info: {info['info'][:100]}...")

Key Endpoints

Endpoint	Description
/sdapi/v1/txt2img	Text to image
/sdapi/v1/img2img	Image to image
/sdapi/v1/extra-single-image	Upscale
/sdapi/v1/sd-models	List models
/sdapi/v1/samplers	List samplers
/sdapi/v1/loras	List LoRAs
/sdapi/v1/options	Get/set options
/sdapi/v1/progress	Generation progress
/sdapi/v1/interrupt	Cancel generation

---

Need custom data extraction or scraping solution? I build production-grade scrapers for any website. Email: [e-mail] | My Apify Actors