iFlow

透過 Mastra 的模型路由器存取 14 個 iFlow 模型。系統會自動使用 IFLOW_API_KEY 環境變數完成身分驗證。

如需進一步了解，請參閱iFlow 文件。

.env

IFLOW_API_KEY=your-api-key

src/mastra/agents/my-agent.ts

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

const agent = new Agent({
  id: 'my-agent',
  name: 'My Agent',
  instructions: 'You are a helpful assistant',
  model: 'iflowcn/deepseek-r1',
})

// 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)
}

資訊

Mastra 使用與 OpenAI 相容的 /chat/completions endpoint。部分 Provider 特有功能可能無法使用，詳情請參閱iFlow 文件。

模型

14 available models

Model	Context	Tools	Reasoning	Image	Audio	Video	Input $/1M	Output $/1M
iflowcn/deepseek-r1	128K						—	—
iflowcn/deepseek-v3	128K						—	—
iflowcn/deepseek-v3.2	128K						—	—
iflowcn/glm-4.6	200K						—	—
iflowcn/kimi-k2	128K						—	—
iflowcn/kimi-k2-0905	256K						—	—
iflowcn/qwen3-235b	128K						—	—
iflowcn/qwen3-235b-a22b-instruct	256K						—	—
iflowcn/qwen3-235b-a22b-thinking-2507	256K						—	—
iflowcn/qwen3-32b	128K						—	—
iflowcn/qwen3-coder-plus	256K						—	—
iflowcn/qwen3-max	256K						—	—
iflowcn/qwen3-max-preview	256K						—	—
iflowcn/qwen3-vl-plus	256K						—	—

進階設定

自訂標頭

src/mastra/agents/my-agent.ts

const agent = new Agent({
  id: 'custom-agent',
  name: 'custom-agent',
  model: {
    url: 'https://apis.iflow.cn/v1',
    id: 'iflowcn/deepseek-r1',
    apiKey: process.env.IFLOW_API_KEY,
    headers: {
      'X-Custom-Header': 'value',
    },
  },
})

動態選擇模型

src/mastra/agents/my-agent.ts

const agent = new Agent({
  id: 'dynamic-agent',
  name: 'Dynamic Agent',
  model: ({ requestContext }) => {
    const useAdvanced = requestContext.task === 'complex'
    return useAdvanced ? 'iflowcn/qwen3-vl-plus' : 'iflowcn/deepseek-r1'
  },
})