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

# Cohere Provider

> Learn how to use Cohere models for text generation, embeddings, and reranking with the AI SDK.

# Cohere Provider

The Cohere provider enables you to use Cohere's language models for text generation, embeddings, and document reranking.

## Installation

<CodeGroup>
  ```bash npm theme={null}
  npm install @ai-sdk/cohere
  ```

  ```bash pnpm theme={null}
  pnpm add @ai-sdk/cohere
  ```

  ```bash yarn theme={null}
  yarn add @ai-sdk/cohere
  ```

  ```bash bun theme={null}
  bun add @ai-sdk/cohere
  ```
</CodeGroup>

## Setup

Get your API key from [Cohere Dashboard](https://dashboard.cohere.com/) and set it as an environment variable:

```bash theme={null}
COHERE_API_KEY=your-api-key
```

## Usage

### Basic Text Generation

```typescript theme={null}
import { cohere } from '@ai-sdk/cohere';
import { generateText } from 'ai';

const { text } = await generateText({
  model: cohere('command-r-plus'),
  prompt: 'Explain natural language processing.',
});
```

### Streaming Responses

```typescript theme={null}
import { cohere } from '@ai-sdk/cohere';
import { streamText } from 'ai';

const result = streamText({
  model: cohere('command-r-plus'),
  prompt: 'Write about language models.',
});

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

### Structured Output

```typescript theme={null}
import { cohere } from '@ai-sdk/cohere';
import { generateObject } from 'ai';
import { z } from 'zod';

const { object } = await generateObject({
  model: cohere('command-r-plus'),
  schema: z.object({
    summary: z.string(),
    keywords: z.array(z.string()),
  }),
  prompt: 'Extract information from this text.',
});
```

## Configuration

### Custom Provider Instance

```typescript theme={null}
import { createCohere } from '@ai-sdk/cohere';

const cohere = createCohere({
  apiKey: process.env.COHERE_API_KEY,
  baseURL: 'https://api.cohere.com/v2',
});
```

### Provider Options

* **apiKey**: Your Cohere API key (defaults to `COHERE_API_KEY` env var)
* **baseURL**: Custom API endpoint URL
* **headers**: Custom headers for requests

## Available Models

### Text Models

* `command-a-03-2025` - Latest Command model
* `command-r-plus` - Most capable model
* `command-r` - Balanced performance
* `command-r7b-12-2024` - Efficient 7B model

### Reasoning Models

* `command-a-reasoning-08-2025` - Reasoning model

## Advanced Features

### Tool Calling

```typescript theme={null}
import { cohere } from '@ai-sdk/cohere';
import { generateText } from 'ai';
import { z } from 'zod';

const result = await generateText({
  model: cohere('command-r-plus'),
  tools: {
    search_docs: {
      description: 'Search documentation',
      parameters: z.object({
        query: z.string(),
      }),
      execute: async ({ query }) => {
        return { results: [] };
      },
    },
  },
  prompt: 'Find information about embeddings.',
});
```

### Reasoning

```typescript theme={null}
import { cohere } from '@ai-sdk/cohere';
import { generateText } from 'ai';

const { text, reasoning } = await generateText({
  model: cohere('command-a-reasoning-08-2025'),
  prompt: 'How many sisters does Alice\'s brother have if Alice has 3 brothers and 2 sisters?',
  providerOptions: {
    cohere: {
      thinking: {
        type: 'enabled',
        tokenBudget: 100,
      },
    },
  },
});

console.log('Reasoning:', reasoning);
console.log('Answer:', text);
```

### Embeddings

```typescript theme={null}
import { cohere } from '@ai-sdk/cohere';
import { embed } from 'ai';

const { embedding } = await embed({
  model: cohere.embedding('embed-english-v3.0'),
  value: 'Document to embed',
  providerOptions: {
    cohere: {
      inputType: 'search_document',
    },
  },
});
```

**Available Models:**

* `embed-english-v3.0` (1024 dimensions)
* `embed-multilingual-v3.0` (1024 dimensions)
* `embed-english-light-v3.0` (384 dimensions)

**Input Types:**

* `search_document` - For documents in a vector database
* `search_query` - For search queries
* `classification` - For classification tasks
* `clustering` - For clustering tasks

### Reranking

```typescript theme={null}
import { cohere } from '@ai-sdk/cohere';
import { rerank } from 'ai';

const documents = [
  'AI is transforming healthcare.',
  'Machine learning powers recommendations.',
  'Cloud computing enables scalability.',
];

const { ranking } = await rerank({
  model: cohere.reranking('rerank-v3.5'),
  documents,
  query: 'artificial intelligence applications',
  topN: 2,
});

console.log(ranking);
// [
//   { originalIndex: 0, score: 0.95, document: 'AI is transforming healthcare.' },
//   { originalIndex: 1, score: 0.87, document: 'Machine learning powers recommendations.' }
// ]
```

**Available Models:**

* `rerank-v3.5` - Latest reranking model
* `rerank-english-v3.0` - English-only
* `rerank-multilingual-v3.0` - Multilingual support

## Model Capabilities

| Model               | Tools | Structured Output | Reasoning |
| ------------------- | ----- | ----------------- | --------- |
| command-a-03-2025   | ✓     | ✓                 | ✗         |
| command-r-plus      | ✓     | ✓                 | ✗         |
| command-r           | ✓     | ✓                 | ✗         |
| command-a-reasoning | ✓     | ✓                 | ✓         |

## Resources

* [Cohere Documentation](https://docs.cohere.com/)
* [AI SDK Documentation](/docs)
* [Cohere API Reference](https://docs.cohere.com/reference)