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

# ![Vivgrid 标志](https://models.dev/logos/vivgrid.svg)Vivgrid

通过 Mastra 的模型路由器访问 19 个 Vivgrid 模型。系统会自动使用 `VIVGRID_API_KEY` 环境变量完成身份验证。

更多信息请参阅[Vivgrid 文档](https://docs.vivgrid.com/models)。

```bash
VIVGRID_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: 'vivgrid/deepseek-v3.2',
})

// 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[Vivgrid 文档](https://docs.vivgrid.com/models)。

## 模型

| Model                                   | Context | Tools | Reasoning | Image | Audio | Video | Input $/1M | Output $/1M |
| --------------------------------------- | ------- | ----- | --------- | ----- | ----- | ----- | ---------- | ----------- |
| `vivgrid/deepseek-v3.2`                 | 128K    |       |           |       |       |       | $0.28      | $0.42       |
| `vivgrid/deepseek-v4-flash`             | 1.0M    |       |           |       |       |       | $0.15      | $0.30       |
| `vivgrid/deepseek-v4-pro`               | 1.0M    |       |           |       |       |       | $0.43      | $0.87       |
| `vivgrid/gemini-3.1-flash-lite-preview` | 1.0M    |       |           |       |       |       | $0.25      | $2          |
| `vivgrid/gemini-3.1-pro-preview`        | 1.0M    |       |           |       |       |       | $2         | $12         |
| `vivgrid/glm-5.2`                       | 1.0M    |       |           |       |       |       | $1         | $4          |
| `vivgrid/gpt-5-mini`                    | 272K    |       |           |       |       |       | $0.25      | $2          |
| `vivgrid/gpt-5.1-codex`                 | 400K    |       |           |       |       |       | $1         | $10         |
| `vivgrid/gpt-5.1-codex-max`             | 400K    |       |           |       |       |       | $1         | $10         |
| `vivgrid/gpt-5.2-codex`                 | 400K    |       |           |       |       |       | $2         | $14         |
| `vivgrid/gpt-5.3-codex`                 | 400K    |       |           |       |       |       | $2         | $14         |
| `vivgrid/gpt-5.4`                       | 400K    |       |           |       |       |       | $3         | $15         |
| `vivgrid/gpt-5.4-mini`                  | 400K    |       |           |       |       |       | $0.75      | $5          |
| `vivgrid/gpt-5.4-nano`                  | 400K    |       |           |       |       |       | $0.20      | $1          |
| `vivgrid/gpt-5.5`                       | 1.1M    |       |           |       |       |       | $5         | $30         |
| `vivgrid/gpt-5.6-luna`                  | 1.1M    |       |           |       |       |       | $1         | $6          |
| `vivgrid/gpt-5.6-sol`                   | 1.1M    |       |           |       |       |       | $5         | $30         |
| `vivgrid/gpt-5.6-terra`                 | 1.1M    |       |           |       |       |       | $3         | $15         |
| `vivgrid/kimi-k3`                       | 1.0M    |       |           |       |       |       | $3         | $15         |

## 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.vivgrid.com/v1',
    id: 'vivgrid/deepseek-v3.2',
    apiKey: process.env.VIVGRID_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 ? 'vivgrid/kimi-k3' : 'vivgrid/deepseek-v3.2'
  },
})
```

## 直接安装 Provider

你也可以直接将此 Provider 安装为独立 package，用它代替 Mastra 模型路由字符串。详情请参阅[package 文档](https://www.npmjs.com/package/@ai-sdk/openai)。

**npm**:

```bash
npm install @ai-sdk/openai
```

**pnpm**:

```bash
pnpm add @ai-sdk/openai
```

**Yarn**:

```bash
yarn add @ai-sdk/openai
```

**Bun**:

```bash
bun add @ai-sdk/openai
```