IO.NET

透過 Mastra 的模型路由器存取 17 個 IO.NET 模型。系統會自動使用 IOINTELLIGENCE_API_KEY 環境變數進行驗證。

詳情請參閱IO.NET 文件。

.env

IOINTELLIGENCE_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: 'io-net/Intel/Qwen3-Coder-480B-A35B-Instruct-int4-mixed-ar',
})

// 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 端點。部分 Provider 專屬功能可能無法使用，詳情請參閱IO.NET 文件。

模型

17 available models

Model	Context	Tools	Reasoning	Image	Audio	Video	Input $/1M	Output $/1M
io-net/deepseek-ai/DeepSeek-R1-0528	128K						$2	$9
io-net/Intel/Qwen3-Coder-480B-A35B-Instruct-int4-mixed-ar	106K						$0.22	$0.95
io-net/meta-llama/Llama-3.2-90B-Vision-Instruct	16K						$0.35	$0.40
io-net/meta-llama/Llama-3.3-70B-Instruct	128K						$0.13	$0.38
io-net/meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8	430K						$0.15	$0.60
io-net/mistralai/Devstral-Small-2505	128K						$0.05	$0.22
io-net/mistralai/Magistral-Small-2506	128K						$0.50	$2
io-net/mistralai/Mistral-Large-Instruct-2411	128K						$2	$6
io-net/mistralai/Mistral-Nemo-Instruct-2407	128K						$0.02	$0.04
io-net/moonshotai/Kimi-K2-Instruct-0905	33K						$0.39	$2
io-net/moonshotai/Kimi-K2-Thinking	33K						$0.55	$2
io-net/openai/gpt-oss-120b	131K						$0.04	$0.40
io-net/openai/gpt-oss-20b	64K						$0.03	$0.14
io-net/Qwen/Qwen2.5-VL-32B-Instruct	32K						$0.05	$0.22
io-net/Qwen/Qwen3-235B-A22B-Thinking-2507	262K						$0.11	$0.60
io-net/Qwen/Qwen3-Next-80B-A3B-Instruct	262K						$0.10	$0.80
io-net/zai-org/GLM-4.6	200K						$0.40	$2

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://api.intelligence.io.solutions/api/v1',
    id: 'io-net/Intel/Qwen3-Coder-480B-A35B-Instruct-int4-mixed-ar',
    apiKey: process.env.IOINTELLIGENCE_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
      ? 'io-net/zai-org/GLM-4.6'
      : 'io-net/Intel/Qwen3-Coder-480B-A35B-Instruct-int4-mixed-ar'
  },
})