> ## Documentation Index
> Fetch the complete documentation index at: https://docs.verlon.ai/llms.txt
> Use this file to discover all available pages before exploring further.
# Overview
> Introduction to the Verlon AI SDK for TypeScript/JavaScript
The Verlon SDK provides a unified interface for AI capabilities across multiple providers. Use one SDK to access chat, image generation, embeddings, text-to-speech, video generation, and OCR.
## Installation
```bash npm theme={null}
npm install @verlon-ai/sdk
```
```bash pnpm theme={null}
pnpm add @verlon-ai/sdk
```
```bash yarn theme={null}
yarn add @verlon-ai/sdk
```
## Quick Start
```typescript theme={null}
import { Verlon } from '@verlon-ai/sdk';
const verlon = new Verlon({
apiKey: process.env.VERLON_API_KEY
});
const response = await verlon.chat({
gateId: 'your-gate-id',
data: {
messages: [
{ role: 'user', content: 'Hello!' }
]
}
});
console.log(response.content);
```
## Configuration
### Initialize the Client
```typescript theme={null}
import { Verlon } from '@verlon-ai/sdk';
const verlon = new Verlon({
apiKey: 'your-api-key', // Required
baseUrl: 'https://custom.api' // Optional, defaults to https://api.verlon.ai
});
```
Never commit API keys to version control. Use environment variables.
### Get Your API Key
1. Sign up at [verlon.ai/signup](https://verlon.ai/signup)
2. Get your API key from the [Dashboard](https://verlon.ai/dashboard)
3. Store it in an environment variable
```bash .env theme={null}
VERLON_API_KEY=your-api-key-here
```
## Available Methods
The SDK provides methods for different AI capabilities:
Text generation with streaming, vision, and function calling
Generate images from text prompts
Create videos from text descriptions
Generate vector embeddings for text
Convert text to speech
Extract text from images
## Response Structure
All SDK methods return a `VerlonResponse` object:
```typescript theme={null}
interface VerlonResponse {
// Common fields
id: string; // Request ID
model: string; // Model used
cost: number; // Request cost in USD
latency: number; // Response time in ms
// Method-specific fields
content?: string; // Chat response text
data?: any[]; // Image/video/embedding data
url?: string; // Audio file URL
text?: string; // OCR extracted text
// Usage statistics
usage?: {
promptTokens: number;
completionTokens: number;
totalTokens: number;
};
}
```
## Error Handling
Wrap SDK calls in try-catch blocks:
```typescript theme={null}
try {
const response = await verlon.chat({
gateId: 'your-gate-id',
data: {
messages: [{ role: 'user', content: 'Hello!' }]
}
});
console.log(response.content);
} catch (error) {
console.error('Error:', error.message);
}
```
### Common Error Messages
| Error | Cause |
| ---------------------- | ---------------------------- |
| `Invalid token` | Check your API key |
| `Gate not found` | Verify gate ID is correct |
| `Rate limit exceeded` | Too many requests, slow down |
| `Insufficient credits` | Add credits to your account |
## TypeScript Support
The SDK is fully typed:
```typescript theme={null}
import { Verlon, ChatRequest, VerlonResponse } from '@verlon-ai/sdk';
const verlon = new Verlon({
apiKey: process.env.VERLON_API_KEY!
});
// TypeScript knows the structure
const response: VerlonResponse = await verlon.chat({
gateId: 'your-gate-id',
data: {
messages: [
{ role: 'user', content: 'Hello!' }
]
}
});
// Autocomplete and type safety
console.log(response.content); // TypeScript knows this exists
console.log(response.cost); // TypeScript knows this is a number
```
## Environment Variables
Best practice for managing credentials:
```typescript theme={null}
// Good ✅
const verlon = new Verlon({
apiKey: process.env.VERLON_API_KEY
});
// Bad ❌
const verlon = new Verlon({
apiKey: 'sk-vrln-hardcoded_key' // Never do this!
});
```
## Next Steps
Explore specific capabilities:
Or learn about core concepts:
How Verlon routes requests to models
Configure gates in the dashboard