Skip to main content

​
Getting Started with Universal AI

The Universal AI endpoint is the core of Eden AI V3, providing a single unified endpoint for all non-LLM AI features.

​
Overview

Instead of calling different endpoints for different features, V3’s Universal AI endpoint handles everything through model strings: 
POST /v3/universal-ai
One endpoint for:
  • Text analysis (moderation, AI detection, embeddings, sentiment)
  • OCR (text extraction, invoice/ID parsing)
  • Image processing (generation, detection, analysis)
  • Translation (document translation)

​
Model String Format

The model string tells the endpoint what feature and provider to use: 
feature/subfeature/provider[/model]
Examples:
  • text/moderation/google
  • ocr/financial_parser/google
  • image/generation/openai/dall-e-3
  • translation/document_translation/deepl

​
Basic Request

import requests

url = "https://api.edenai.run/v3/universal-ai"
headers = {
    "Authorization": "Bearer YOUR_API_KEY",
    "Content-Type": "application/json"
}

payload = {
    "model": "text/moderation/openai",
    "input": {
        "text": "This is sample text to moderate"
    }
}

response = requests.post(url, headers=headers, json=payload)
result = response.json()
print(result)

​
Response Format

All Universal AI responses follow the same structure: 
{
  "status": "success",
  "cost": 0.0001,
  "provider": "openai",
  "feature": "text",
  "subfeature": "moderation",
  "output": {
    // Feature-specific output
  }
}

​
Input Formats

The input field varies based on the feature:

​
Text-Based Features

{
  "model": "text/moderation/google",
  "input": {
    "text": "Text to moderate"
  }
}

​
File-Based Features (UUID)

{
  "model": "ocr/financial_parser/google",
  "input": {
    "file": "550e8400-e29b-41d4-a716-446655440000"
  }
}

​
File-Based Features (URL)

{
  "model": "image/object_detection/google",
  "input": {
    "file": "https://example.com/image.jpg"
  }
}

​
Common Use Cases

​
Text Moderation

import requests

url = "https://api.edenai.run/v3/universal-ai"
headers = {
    "Authorization": "Bearer YOUR_API_KEY",
    "Content-Type": "application/json"
}

payload = {
    "model": "text/moderation/openai",
    "input": {"text": "Content to moderate"}
}
    
response = requests.post(url, headers=headers, json=payload)
result = response.json()
    
if result["output"]["nsfw_likelihood"] > 3:
    print("Content flagged as inappropriate")

​
OCR Text Extraction

import requests

url = "https://api.edenai.run/v3/universal-ai"
headers = {
    "Authorization": "Bearer YOUR_API_KEY",
    "Content-Type": "application/json"
}

# First upload the file
upload_response = requests.post(
    "https://api.edenai.run/v3/upload",
    headers={"Authorization": "Bearer YOUR_API_KEY"},
    files={"file": open("document.pdf", "rb")}
)
file_id = upload_response.json()["file_id"]

# Then use it in Universal AI
payload = {
    "model": "ocr/financial_parser/google",
    "input": {"file": file_id, "language": "en"}
}
    
response = requests.post(url, headers=headers, json=payload)
result = response.json()
print(result["output"]["extracted_data"])

​
Next Steps