ai-model-router
Cloudflare Workers MCP server: ai-model-router
我该使用它吗
质量与安全性
B
基于对工具定义和协议合规性的自动分析。
上下文开销
~269token 数（工具定义）
~725 B典型响应大小
对注意力的影响极小（占 128k 上下文窗口的 0.21%）
这是每次将服务器的工具加载到模型上下文窗口时所消耗的大致 token 数。数值越高，可用于其他任务的注意力就越少。
安装
一键安装
将以下内容添加到你的 `claude_desktop_config.json` 文件中：
{
"mcpServers": {
"ai-model-router": {
"url": "https://api.lazy-mac.com/ai-model-router/mcp"
}
}
}远程端点
https://api.lazy-mac.com/ai-model-router/mcpstreamable-http它能做什么
工具清单
工具（3）
🟢 只读🟡 写入🔴 删除⚪ 未知
🟢route_model(prompt, task_type, max_cost_usd, prefer_provider, prefer_tier, ...)
Select the optimal AI model (GPT-4, Claude, Gemini) for a task based on type, cost, and provider preferences.
输入模式
{
"type": "object",
"properties": {
"prompt": {
"type": "string",
"description": "Task description for auto-classification"
},
"task_type": {
"type": "string",
"enum": [
"code",
"analysis",
"creative",
"summarize",
"classify",
"reason",
"vision",
"chat",
"general"
]
},
"max_cost_usd": {
"type": "number",
"description": "Max budget in USD per call"
},
"prefer_provider": {
"type": "string",
"enum": [
"openai",
"anthropic",
"google"
]
},
"prefer_tier": {
"type": "string",
"enum": [
"economy",
"standard",
"premium"
]
},
"input_tokens": {
"type": "integer"
},
"output_tokens": {
"type": "integer"
}
}
}⚪compare_models(prompt, task_type, input_tokens, output_tokens)
Compare cost and task fit across all supported AI models for a given prompt or token count.
输入模式
{
"type": "object",
"properties": {
"prompt": {
"type": "string"
},
"task_type": {
"type": "string"
},
"input_tokens": {
"type": "integer"
},
"output_tokens": {
"type": "integer"
}
}
}🟢list_models
List all supported AI models with pricing and capabilities.
输入模式
{
"type": "object",
"properties": {}
}社区
证据
最近观测
已验证未记录版本3 个工具
已验证未记录版本3 个工具
已验证未记录版本3 个工具