> ## Documentation Index
> Fetch the complete documentation index at: https://cli.terramind.com/llms.txt
> Use this file to discover all available pages before exploring further.

# AI SDK Provider Overview

> Use Terramind as a provider with the Vercel AI SDK

## Introduction

Terramind can be used as an AI provider with the [Vercel AI SDK](https://sdk.vercel.ai/), giving you access to multiple LLM providers through a unified, easy-to-use interface.

This is perfect for building AI-powered applications without managing multiple API keys or dealing with different provider APIs.

## Why Use Terramind as a Provider?

<CardGroup cols={2}>
  <Card title="Unified Interface" icon="plug">
    Access Claude, GPT, Gemini, and more through one consistent API
  </Card>

  <Card title="Simplified Auth" icon="key">
    One API key for all models - no need to manage multiple provider keys
  </Card>

  <Card title="Built-in Tracking" icon="chart-line">
    Automatic usage tracking and cost monitoring
  </Card>

  <Card title="Vercel AI SDK Compatible" icon="code">
    Drop-in replacement that works with all AI SDK features
  </Card>
</CardGroup>

## Features

### Full AI SDK Support

Terramind supports all major Vercel AI SDK features:

* ✅ **Text Generation** - `generateText()`, `streamText()`
* ✅ **Object Generation** - `generateObject()`, `streamObject()`
* ✅ **Tool Calling** - Full tool/function calling support
* ✅ **Streaming** - Real-time streaming responses
* ✅ **Multi-modal** - Text, images, and more
* ✅ **Embeddings** - Generate text embeddings

### Multiple Models

Access industry-leading models:

* **Claude**: Sonnet 4.5, Sonnet 4, Opus 4.1, Haiku 4.5
* **GPT**: GPT-5, GPT-5 Codex
* **Chinese Models**: GLM-4.6, Kimi K2, Qwen3 Coder
* **Specialized**: Grok Code, Big Pickle

## Installation

Install both Terramind and the Vercel AI SDK:

<CodeGroup>
  ```bash npm theme={null}
  npm install terramind ai
  ```

  ```bash pnpm theme={null}
  pnpm add terramind ai
  ```

  ```bash bun theme={null}
  bun add terramind ai
  ```
</CodeGroup>

## Quick Start

Here's a simple example to get started:

```typescript theme={null}
import { createTerramind } from "terramind/provider"
import { generateText } from "ai"

const terramind = createTerramind({
  apiKey: process.env.TERRAMIND_API_KEY,
})

const result = await generateText({
  model: terramind("claude-sonnet-4-5"),
  prompt: "Write a haiku about coding",
})

console.log(result.text)
```

## Basic Usage Patterns

### Text Generation

Generate text responses:

```typescript theme={null}
const result = await generateText({
  model: terramind("claude-sonnet-4-5"),
  prompt: "Explain React hooks in simple terms",
})

console.log(result.text)
```

### Streaming Responses

Stream responses in real-time:

```typescript theme={null}
const result = await streamText({
  model: terramind("gpt-5"),
  prompt: "Write a short story about AI",
})

for await (const chunk of result.textStream) {
  process.stdout.write(chunk)
}
```

### Object Generation

Generate structured data:

```typescript theme={null}
import { generateObject } from "ai"
import { z } from "zod"

const result = await generateObject({
  model: terramind("claude-sonnet-4-5"),
  schema: z.object({
    recipe: z.string(),
    ingredients: z.array(z.string()),
    steps: z.array(z.string()),
  }),
  prompt: "Generate a recipe for chocolate chip cookies",
})

console.log(result.object)
```

### Tool Calling

Use AI with tools/functions:

```typescript theme={null}
import { generateText, tool } from "ai"
import { z } from "zod"

const result = await generateText({
  model: terramind("claude-sonnet-4-5"),
  prompt: "What's the weather in San Francisco?",
  tools: {
    getWeather: tool({
      description: "Get the weather for a location",
      parameters: z.object({
        location: z.string(),
      }),
      execute: async ({ location }) => {
        // Your weather API call here
        return { temperature: 72, condition: "sunny" }
      },
    }),
  },
})
```

## Configuration

### Create Provider Instance

```typescript theme={null}
const terramind = createTerramind({
  apiKey: process.env.TERRAMIND_API_KEY,
  baseURL: "https://api.terramind.com", // Optional
})
```

### Environment Variables

Set your API key as an environment variable:

```bash theme={null}
export TERRAMIND_API_KEY=your_api_key_here
```

Then the provider will automatically use it:

```typescript theme={null}
const terramind = createTerramind() // Uses TERRAMIND_API_KEY
```

## Error Handling

Handle errors gracefully:

```typescript theme={null}
try {
  const result = await generateText({
    model: terramind("claude-sonnet-4-5"),
    prompt: "Hello!",
  })
  console.log(result.text)
} catch (error) {
  if (error.name === "AI_APICallError") {
    console.error("API call failed:", error.message)
  } else if (error.name === "AI_InvalidPromptError") {
    console.error("Invalid prompt:", error.message)
  } else {
    console.error("Unexpected error:", error)
  }
}
```

## TypeScript Support

Terramind has full TypeScript support with type inference:

```typescript theme={null}
import { createTerramind } from "terramind/provider"
import { generateText } from "ai"

const terramind = createTerramind({
  apiKey: process.env.TERRAMIND_API_KEY!,
})

// TypeScript knows the return type
const result = await generateText({
  model: terramind("claude-sonnet-4-5"),
  prompt: "Hello",
})

// result.text is string
// result.usage is { promptTokens: number, completionTokens: number }
```

## Next Steps

<CardGroup cols={2}>
  <Card title="Usage Examples" icon="code" href="/provider/usage">
    See detailed code examples and patterns
  </Card>

  <Card title="Available Models" icon="list" href="/provider/models">
    Browse all available models
  </Card>
</CardGroup>