How to Configure AI Services

The AI Services section in Geta.ai allows you to configure and manage the AI models used across your account.

These AI models power:

• 📨 Smart Reply Assistant (Inbox Module)

• 📝 AI Template Generator

• 🤖 Other AI-driven features across the platform

By selecting the right model, you control:

• Response quality

• Performance speed

• Cost per usage

• AI provider preference

Where AI Services Are Used

The selected AI model is applied to:

1️⃣ Smart Reply Assistant (Inbox)

Automatically generates AI-powered replies for customer conversations.

2️⃣ AI Template Generator

Helps create WhatsApp templates using AI.

⚠️ Important: The model selected in AI Services directly affects cost and output quality in these modules.

How to Configure Smart AI LLM

Step 1: Navigate to AI Services

1. Log in to your Geta.ai account

2. Go to Settings

3. Click on AI Services

4. The AI Services Configuration page opens

Step 2: Select Smart AI LLM

1. Click on Smart AI LLM

2. You will see:

   • 100+ models

   • Multiple providers (OpenAI, Google, Anthropic, DeepSeek, etc.)

   • Pricing per model

Step 3: Filter and Select a Model

1. Use Filter by Provider dropdown

   • Choose from:

     • All Providers

     • OpenAI

     • Google

     • Anthropic

     • DeepSeek

     • etc.

2. Browse the model cards

Each model displays:

• ✅ Model Name

• 💲 Input Price (per million tokens)

• 💲 Output Price (per million tokens)

• 🔘 Select Model button

3. Click Select Model

4. Click Save

Your selected model will now power:

• Inbox Smart Reply

• AI Template Generator

Understanding AI Model Pricing

Each AI model has:

• Input Price → Cost for tokens you send to the model

• Output Price → Cost for tokens the model generates

Tokens represent chunks of text (words, punctuation, characters).

💰 Cost Calculation Formula

The total AI cost is calculated using this formula:

Total Cost =
(Input Tokens / 1,000,000 × Input Price)
+
(Output Tokens / 1,000,000 × Output Price)

📊 Example Calculation

Let’s assume the selected model has:

• Input Price = $1.25 per 1M tokens

• Output Price = $10.00 per 1M tokens

If your automation used:

• 50,000 input tokens

• 20,000 output tokens

Step 1 – Input Cost

(50,000 / 1,000,000) × 1.25
= 0.05 × 1.25
= $0.0625

Step 2 – Output Cost

(20,000 / 1,000,000) × 10.00
= 0.02 × 10.00
= $0.20

Step 3 – Total Cost

$0.0625 + $0.20 = $0.2625

✅ Total AI cost = $0.2625

📌 Quick Reference Example (1 Million Tokens Split 50/50)

If total usage = 1,000,000 tokens:

⚠️ Important Notes

• If no model or API key is selected, the system automatically uses the default model.

• Costs are deducted from your available wallet balance.

• Higher-quality models typically have higher output pricing.

• Output tokens usually cost more than input tokens.

How to Configure Your Own OpenAI API Key

If you prefer to use your own OpenAI account:

Step 1: Switch to Custom API

1. Go to AI Services

2. Click on Configure Your Own OpenAI Key

Step 2: API Configuration

1. Enter your OpenAI API Key

2. Click Select Model

3. Choose your preferred OpenAI model

4. Click Save

Now:

• Your OpenAI account will be billed directly

• Geta.ai will not deduct wallet balance for AI usage

When Should You Use Smart AI LLM vs Your Own API?

Best Practices for Cost Optimization

• Use smaller models for template generation

• Use advanced models only when needed

• Monitor token-heavy automations

• Keep prompts concise

• Avoid unnecessary repeated calls

Summary

The AI Services module in Geta.ai allows you to:

• Choose from 100+ AI models

• Filter by provider

• Understand pricing before selection

• Control how AI works in Inbox & Template Generator

• Calculate cost transparently using token-based billing

By understanding the pricing formula and token usage, you can optimize both performance and cost efficiently.