Generate Images with the Stability AI Platform API

1. 1
   Get Your API Key

   First, you need an API key. Navigate to the Stability AI Platform dashboard at platform.stability.ai, create an account or log in, and go to the API Keys section to generate your key. For security, set it as an environment variable named 'STABILITY_API_KEY'.
2. 2
   Set Up Your Python Environment

   You'll need Python installed. The only external library required is `requests` for making HTTP calls. It's a standard library for this purpose and you can install it using pip.
3. 3
   Craft the API Request Script

   Create a Python script to call the Stability AI Platform API. The script will send a POST request to the Stable Diffusion 3 endpoint. You'll specify your prompt, the model (e.g., 'sd3-medium'), and the desired output format in the request body.
4. 4
   Save and View the Generated Image

   After making the request, check if the status code is 200 (OK). If it is, the response body contains the image data. Write this binary data to a file (e.g., 'output.png') and then open it to see your generated image.

import os
import requests
import sys

# --- Configuration ---
# Get API key from environment variable
API_KEY = os.getenv("STABILITY_API_KEY")
API_HOST = 'https://api.stability.ai'
OUTPUT_FILE = "output.png"

# --- Validation ---
if not API_KEY:
    sys.exit("Error: STABILITY_API_KEY environment variable not set. Please get your key from https://platform.stability.ai/account/keys")

# --- API Request ---
print("Calling Stability AI Platform API...")

response = requests.post(
    f"{API_HOST}/v2beta/stable-image/generate/sd3",
    headers={
        "authorization": f"Bearer {API_KEY}",
        "accept": "image/*"  # Request image binary in response
    },
    # Use 'files' for multipart/form-data, required by this endpoint
    files={"none": ''},
    # 'data' holds the parameters for generation
    data={
        "prompt": "A beautiful watercolor painting of a giant turtle swimming in a coral reef",
        "model": "sd3-medium", # Specify the SD3 Medium model
        "output_format": "png",
        "aspect_ratio": "16:9"
    },
)

# --- Response Handling ---
if response.status_code == 200:
    try:
        with open(OUTPUT_FILE, 'wb') as f:
            f.write(response.content)
        print(f"\nSuccess! Image saved to {os.path.abspath(OUTPUT_FILE)}")
    except IOError as e:
        sys.exit(f"Error: Could not write file to disk: {e}")
else:
    print(f"\nError: API call failed with status code {response.status_code}")
    try:
        # Print the error details if the response is JSON
        print(response.json())
    except requests.exceptions.JSONDecodeError:
        # Print the raw text if it's not JSON
        print(response.text)
    sys.exit(1)