> Discover all available pages from the documentation index: https://mastra.zisheng.pro/llms.txt

# ![Sarvam AI 标志](https://models.dev/logos/sarvam.svg)Sarvam AI

通过 Mastra 的模型路由器访问 2 个 Sarvam AI 模型。系统会自动使用 `SARVAM_API_KEY` 环境变量完成身份验证。

更多信息请参阅[Sarvam AI 文档](https://docs.sarvam.ai)。

```bash
SARVAM_API_KEY=your-api-key
```

```typescript
import { Agent } from '@mastra/core/agent'

const agent = new Agent({
  id: 'my-agent',
  name: 'My Agent',
  instructions: 'You are a helpful assistant',
  model: 'sarvam/sarvam-105b',
})

// Generate a response
const response = await agent.generate('Hello!')

// Stream a response
const stream = await agent.stream('Tell me a story')
for await (const chunk of stream) {
  console.log(chunk)
}
```

> **信息:** Mastra 使用兼容 OpenAI 的 `/chat/completions` endpoint。部分 Provider 特有功能可能不可用，详情请参阅[Sarvam AI 文档](https://docs.sarvam.ai)。

## 模型

| Model                | Context | Tools | Reasoning | Image | Audio | Video | Input $/1M | Output $/1M |
| -------------------- | ------- | ----- | --------- | ----- | ----- | ----- | ---------- | ----------- |
| `sarvam/sarvam-105b` | 131K    |       |           |       |       |       | —          | —           |
| `sarvam/sarvam-30b`  | 66K     |       |           |       |       |       | —          | —           |

## 高级配置

### 自定义请求头

```typescript
const agent = new Agent({
  id: 'custom-agent',
  name: 'custom-agent',
  model: {
    url: 'https://api.sarvam.ai/v1',
    id: 'sarvam/sarvam-105b',
    apiKey: process.env.SARVAM_API_KEY,
    headers: {
      'X-Custom-Header': 'value',
    },
  },
})
```

### 动态选择模型

```typescript
const agent = new Agent({
  id: 'dynamic-agent',
  name: 'Dynamic Agent',
  model: ({ requestContext }) => {
    const useAdvanced = requestContext.task === 'complex'
    return useAdvanced ? 'sarvam/sarvam-30b' : 'sarvam/sarvam-105b'
  },
})
```