Quick Start Guide

This guide walks you through creating your first AI-generated product visualization in just a few minutes.

What We'll Build

By the end of this tutorial, you'll have:

• ✅ Created a product from a template
• ✅ Generated a LoRA model for AI predictions
• ✅ Produced a stunning AI image of your product

Prerequisites

Before starting, ensure you have:

• 🔑 API token from your Veez.ai administrator
• 🖼️ Product image (512x512+ recommended, JPG/PNG format)
• 💻 Terminal or API client (we'll use cURL)

Step 1: Set Up Your Environment

First, save your API token as an environment variable for easy use:

export VEEZAI_API_TOKEN="your_api_token_here"

Test your connection:

curl -H "Authorization: Bearer $VEEZAI_API_TOKEN" \
  https://app.veez.ai/api/template/ | jq '.[0]'

You should see template data returned. If you get an error, verify your token.

Step 2: Choose a Template

List available templates and find one that supports LoRA:

curl -H "Authorization: Bearer $VEEZAI_API_TOKEN" \
  https://app.veez.ai/api/template/ | jq '.[] | select(.lora == "available")'

Let's use template ID 279 for this example. Get its details:

curl -H "Authorization: Bearer $VEEZAI_API_TOKEN" \
  https://app.veez.ai/api/template/279 | jq '.'

Note the required textures in the response. You'll see something like:

{
  "textures": [
    {
      "key": "TEX01",
      "name": "Design Pack", 
      "required": 1
    }
  ]
}

Step 3: Prepare Your Image

For this example, ensure your product image is:

• Format: JPG or PNG
• Size: At least 512x512 pixels
• Content: Clear product design or logo
• Quality: High resolution, good contrast

Save your image as my-product-design.jpg in your current directory.

Step 4: Create Your Product

Create a product using the template and your image:

curl -X POST https://app.veez.ai/api/product/ \
  -H "Authorization: Bearer $VEEZAI_API_TOKEN" \
  -F "template_id=279" \
  -F "name=My First Product" \
  -F "description=Created with Veez.ai Quick Start Guide" \
  -F "texture[0]=@my-product-design.jpg" \
  -F "texture[0][key]=TEX01"

Save the product ID from the response:

# Example response:
{
  "id": "abc123def456ghi789",
  "template_id": 279,
  "name": "My First Product",
  "lora_enabled": 0
}

Store the product ID:

export PRODUCT_ID="abc123def456ghi789"

Step 5: Generate LoRA Model

Now generate the LoRA model that enables AI predictions:

curl -X POST https://app.veez.ai/api/product/$PRODUCT_ID/generate-lora \
  -H "Authorization: Bearer $VEEZAI_API_TOKEN"

You should see:

{
  "result": "LoRA creation in progress"
}

Step 6: Wait for LoRA Completion

LoRA generation takes 10-30 minutes. Check the status:

# Check every few minutes
curl -H "Authorization: Bearer $VEEZAI_API_TOKEN" \
  https://app.veez.ai/api/product/$PRODUCT_ID | jq '.lora_enabled'

When it returns 1, your LoRA is ready!

Automated Waiting Script

Or use this script to wait automatically:

#!/bin/bash
echo "Waiting for LoRA to complete..."
while true; do
  status=$(curl -s -H "Authorization: Bearer $VEEZAI_API_TOKEN" \
    https://app.veez.ai/api/product/$PRODUCT_ID | jq -r '.lora_enabled')
  
  if [ "$status" = "1" ]; then
    echo "✅ LoRA is ready!"
    break
  else
    echo "⏳ Still processing... (checking again in 60 seconds)"
    sleep 60
  fi
done

Step 7: Create Your First Prediction

Once LoRA is ready, create an AI-generated image:

curl -X POST https://app.veez.ai/api/prediction/ \
  -H "Authorization: Bearer $VEEZAI_API_TOKEN" \
  -F "product_id=$PRODUCT_ID" \
  -F "prompt=A premium product on a marble countertop in a modern kitchen, soft natural lighting, professional photography style" \
  -F "aspect_ratio=16:9"

Save the prediction ID from the response:

export PREDICTION_ID="prediction_id_from_response"

Step 8: Download Your Image

Wait about 60 seconds, then check your prediction:

curl -H "Authorization: Bearer $VEEZAI_API_TOKEN" \
  https://app.veez.ai/api/prediction/$PREDICTION_ID | jq '.images[]'

You'll get image URLs like:

"https://apptest.veez.ai/record/xyz/prediction/out-0.png"

Download your image:

curl -o my-first-prediction.png "https://apptest.veez.ai/record/xyz/prediction/out-0.png"

🎉 Congratulations!

You've successfully:

1. ✅ Created a product from a 3D template
2. ✅ Generated a LoRA model for AI capabilities
3. ✅ Produced an AI image with your product in a custom scene

Complete Script

Here's the entire workflow in one script:

#!/bin/bash

# Configuration
export VEEZAI_API_TOKEN="your_token_here"
TEMPLATE_ID="279"
PRODUCT_NAME="Quick Start Product"
IMAGE_FILE="my-product-design.jpg"

echo "🚀 Starting Veez.ai Quick Start..."

# 1. Create product
echo "📦 Creating product..."
product_response=$(curl -s -X POST https://app.veez.ai/api/product/ \
  -H "Authorization: Bearer $VEEZAI_API_TOKEN" \
  -F "template_id=$TEMPLATE_ID" \
  -F "name=$PRODUCT_NAME" \
  -F "texture[0]=@$IMAGE_FILE" \
  -F "texture[0][key]=TEX01")

PRODUCT_ID=$(echo $product_response | jq -r '.id')
echo "✅ Product created: $PRODUCT_ID"

# 2. Generate LoRA
echo "🧠 Starting LoRA generation..."
curl -s -X POST https://app.veez.ai/api/product/$PRODUCT_ID/generate-lora \
  -H "Authorization: Bearer $VEEZAI_API_TOKEN" > /dev/null

# 3. Wait for LoRA
echo "⏳ Waiting for LoRA completion..."
while true; do
  status=$(curl -s -H "Authorization: Bearer $VEEZAI_API_TOKEN" \
    https://app.veez.ai/api/product/$PRODUCT_ID | jq -r '.lora_enabled')
  
  if [ "$status" = "1" ]; then
    echo "✅ LoRA ready!"
    break
  else
    echo "   Still processing..."
    sleep 60
  fi
done

# 4. Create prediction
echo "🎨 Generating AI image..."
prediction_response=$(curl -s -X POST https://app.veez.ai/api/prediction/ \
  -H "Authorization: Bearer $VEEZAI_API_TOKEN" \
  -F "product_id=$PRODUCT_ID" \
  -F "prompt=A premium product on a marble surface, professional studio lighting" \
  -F "aspect_ratio=1:1")

PREDICTION_ID=$(echo $prediction_response | jq -r '.id')

# 5. Wait for image generation
echo "⏳ Generating image..."
sleep 60

# 6. Download result
image_url=$(curl -s -H "Authorization: Bearer $VEEZAI_API_TOKEN" \
  https://app.veez.ai/api/prediction/$PREDICTION_ID | jq -r '.images[0]')

if [ "$image_url" != "null" ]; then
  curl -o "result-$(date +%s).png" "$image_url"
  echo "🎉 Success! Image downloaded."
else
  echo "⚠️  Image still processing. Check again in a few minutes:"
  echo "   curl -H \"Authorization: Bearer \$VEEZAI_API_TOKEN\" https://app.veez.ai/api/prediction/$PREDICTION_ID"
fi

Next Steps

Now that you've mastered the basics:

1. Explore Advanced Workflows - Learn pro techniques
2. API Reference - Dive deeper into each endpoint
3. Experiment with different prompts - Try various environments and styles
4. Scale up - Create multiple products and batch generate images

Troubleshooting

Common Issues

401 Unauthorized

• Check your API token is correct
• Ensure Bearer prefix is included

Template doesn't support LoRA

• Use jq '.[] | select(.lora == "available")' to find compatible templates

Image upload fails

• Verify image format (JPG/PNG only)
• Check file size (under 10MB)
• Ensure file path is correct

LoRA takes too long

• Normal processing time is 10-30 minutes
• Check status periodically, don't poll too frequently

Need more help? Check the API Reference or contact support!