ProductAI Integration

ProductAI.photo is an AI-powered service that generates professional product photos from existing images. It enables e-commerce businesses, marketers, and designers to create studio-quality product photography without hiring photographers.

Quick Start

1. Get Your API Key

Visit ProductAI Studio → API Access → Copy your API key

2. Run Setup

cd ~/.openclaw/workspace/productai
scripts/setup.py
# Paste your API key when prompted

3. Generate Images

# Generate product photo with custom background
scripts/generate_photo.py 
  --image https://example.com/product.jpg 
  --prompt "modern living room with natural lighting" 
  --output result.png

# Use multiple reference images (nanobanana/seedream support 2 images)
scripts/generate_photo.py 
  --image https://example.com/product1.jpg https://example.com/product2.jpg 
  --prompt "Put the first image on top of the second image" 
  --output result.png

# High quality with Nano Banana Pro
scripts/generate_photo.py 
  --image https://example.com/product.jpg 
  --prompt "white studio background" 
  --model nanobananapro 
  --output hq.png

# Upscale an image (20 tokens)
scripts/upscale_image.py 
  --image https://example.com/photo.jpg 
  --output upscaled.png

Core Capabilities

Photo Generation (/api/generate) - Background replacement with AI-generated scenes - Product placement in realistic environments - Multi-image compositing (up to 2 reference images with certain models) - Custom prompts for full creative control

Image Upscaling (/api/upscale) - Professional AI upscaling (Magnific Precision Upscale) - Preserves image details without distortion - 20 tokens per upscale

Models Available - gpt-low — GPT Low Quality (2 tokens) - gpt-medium — GPT Medium Quality (3 tokens) - gpt-high — GPT High Quality (8 tokens) - kontext-pro — Kontext Pro (3 tokens) - nanobanana — Nano Banana (3 tokens) — DEFAULT - nanobananapro — Nano Banana Pro (8 tokens) - seedream — Seedream (3 tokens)

Multi-Image Support: nanobanana, nanobananapro, and seedream support up to 2 reference images for advanced compositing.

Configuration

API Setup

Step 1: Get Your API Key

1. Visit ProductAI Studio
2. Navigate to API Access section
3. Click Generate API Key or copy existing key

Step 2: Run Setup Script

cd ~/.openclaw/workspace/productai
scripts/setup.py

This will interactively create config.json with your credentials:

{
  "api_key": "your-api-key-here",
  "api_endpoint": "https://api.productai.photo/v1",
  "default_model": "nanobanana",
  "default_resolution": "1024x1024",
  "plan": "standard"
}

The config file is automatically secured with 600 permissions.

Installation

The integration scripts require Python 3.7+ with these dependencies:

pip install requests pillow

All dependencies are handled automatically by the scripts.

Usage Examples

E-commerce Product Photos

Generate clean product shots for online stores:

scripts/generate_photo.py 
  --image https://example.com/raw-product.jpg 
  --prompt "white studio background with soft shadows" 
  --output store-listing.png

Marketing Campaign Visuals

Create lifestyle shots for advertising:

scripts/generate_photo.py 
  --image https://example.com/bottle.jpg 
  --prompt "outdoor picnic scene with blanket and basket, golden hour lighting" 
  --output campaign-hero.png 
  --model kontext-pro

Multi-Image Compositing

Combine multiple products into one scene:

scripts/generate_photo.py 
  --image https://example.com/product1.jpg https://example.com/product2.jpg 
  --prompt "Place the lipstick on top of the cosmetics box" 
  --model nanobanana 
  --output composite.png

High-Quality Output

Use Nano Banana Pro for premium results:

scripts/generate_photo.py 
  --image https://example.com/product.jpg 
  --prompt "luxury marble countertop setting with morning light" 
  --model nanobananapro 
  --output premium.png

Image Upscaling

Upscale images for print or high-res displays:

scripts/upscale_image.py 
  --image https://example.com/product.jpg 
  --output upscaled.png

Async Workflow (No Wait)

Start generation and check later:

# Start job (prints job ID)
scripts/generate_photo.py 
  --image https://example.com/product.jpg 
  --prompt "studio background" 
  --no-wait

# Output: Job created: 12345

# Later, resume and download
scripts/generate_photo.py 
  --job-id 12345 
  --output result.png

API Reference

See API.md for complete endpoint documentation, request/response formats, and authentication details.

Pricing & Token Costs

ProductAI uses a token-based pricing system:

Operation	Token Cost
GPT Low Quality Generation	2 tokens
GPT Medium Quality Generation	3 tokens
GPT High Quality Generation	8 tokens
Kontext Pro	3 tokens
Nano Banana Pro	8 tokens
Nano Banana	3 tokens
Seedream	3 tokens
Magnific Precision Upscale	20 tokens

Subscription Plans:

Visit ProductAI.photo for current plans and token packages.

Rate Limits: - 15 requests per minute - Daily limits based on subscription plan

Note: Tokens are deducted when each operation starts (not on completion).

Troubleshooting

API Key Issues - Verify config.json exists: cat ~/.openclaw/workspace/productai/config.json - Check API key in ProductAI Studio → API Access - Regenerate key if needed

OUT_OF_TOKENS Error - Check your token balance in ProductAI Studio - Purchase more tokens or upgrade plan - Remember: tokens are deducted when jobs start, not on completion

Rate Limit (429) - Maximum 15 requests per minute - Wait before retrying - Use --no-wait for batch jobs and check status later

Job Failed (ERROR status) - Check image URL is accessible and under 10MB - Verify image format (PNG, JPG, or WebP only) - Check prompt length and content - Try a different model

Multi-Image Not Working - Only nanobanana, nanobananapro, and seedream support multiple images - Maximum 2 reference images - Use array format: --image url1 url2

Support

• Website: https://www.productai.photo
• Documentation: See references/API.md
• Contact: [email protected] (or team contact when available)

Advanced Topics

For detailed API specifications, authentication flows, webhook integration, and batch processing patterns, see API.md.