Image Generation API (Experimental)

We're introducing a new experimental Image Generation API that lets you create images from text prompts through the same secure, metered proxy you use for text, vision, and PDF.

#

What you get

• Works with OpenAI (gpt-image-1, dall-e-3, dall-e-2) and Google AI (imagen-3.0-generate-002).
• Simple endpoint: POST /v1/image (or /v1/image/{projectId}).
• Supports size (e.g., 1024x1024) or aspectRatio (e.g., 16:9), n images, seed, and provider-specific options.
• Responses include base64-encoded images plus provider metadata and warnings when available.

#

Quick start

Send a request with your project auth headers:

curl -X POST \
  https://api.proxed.ai/v1/image/{your-project-id} \
  -H "Authorization: Bearer {your-partial-api-key}.{your-device-token-or-test-key}" \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "A retro-futuristic city skyline at dusk",
    "model": "gpt-image-1",
    "size": "1024x1024"
  }'

#

Docs

See the full API reference: /docs/api-reference/structured-image

#

Pricing

• GPT Image 1 (per image): low $0.011–$0.016, medium $0.042–$0.063, high $0.167–$0.25 depending on size.
• DALL·E 3 (per image): standard $0.04–$0.08, HD $0.08–$0.12 depending on size.
• DALL·E 2 (per image): $0.016–$0.02 depending on size.