TANDIU AI infrastructure data

Read-only LLM API prices, benchmarks, price index, GPU self-hosting sizing and cloud agent data.

我该使用它吗

质量与安全性

B
描述质量
81%
模式完整度
70%
命名质量
91%
投毒风险
100%
权限匹配度
100%
协议合规性
100%

发现(3)

  • LOWTool 'cheapest_model_for_score' description lacks action verb在 cheapest_model_for_score 中
  • LOWTool 'get_price_index' description lacks action verb在 get_price_index 中
  • LOWTool 'list_cloud_agents' description lacks action verb在 list_cloud_agents 中

基于对工具定义和协议合规性的自动分析。

上下文开销

~1,217token 数(工具定义)
~808 B典型响应大小
对注意力有中等影响(占 128k 上下文窗口的 0.95%)

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": {
    "data": {
      "url": "https://www.tandiu.cn/mcp"
    }
  }
}

远程端点

https://www.tandiu.cn/mcpstreamable-http

它能做什么

工具清单

工具(9)

🟢 只读🟡 写入🔴 删除⚪ 未知
🟢about_tandiu_data

What data TANDIU publishes, refresh cadence, price basis and how to cite. Call first.

输入模式

{
  "type": "object",
  "properties": {},
  "required": [],
  "additionalProperties": false
}
🟢list_llm_prices(country, vendor, query, sort_by, limit)

Official API list prices (USD per 1M tokens) with Artificial Analysis, LiveBench and Epoch AI scores for major US and Chinese LLMs. Filter and sort.

输入模式

{
  "type": "object",
  "properties": {
    "country": {
      "type": "string",
      "description": "US, CN or another ISO code"
    },
    "vendor": {
      "type": "string"
    },
    "query": {
      "type": "string",
      "description": "substring of model name"
    },
    "sort_by": {
      "type": "string",
      "enum": [
        "livebench",
        "aa_index",
        "blended_price",
        "input_price",
        "output_price"
      ]
    },
    "limit": {
      "type": "integer",
      "minimum": 1,
      "maximum": 100
    }
  },
  "required": [],
  "additionalProperties": false
}
🟢cheapest_model_for_score(metric, min_score, min_percent_of_top, country, limit)

Cheapest models whose benchmark score meets a threshold (absolute or percent of the best score).

输入模式

{
  "type": "object",
  "properties": {
    "metric": {
      "type": "string",
      "enum": [
        "aa_index",
        "livebench"
      ]
    },
    "min_score": {
      "type": "number"
    },
    "min_percent_of_top": {
      "type": "number",
      "minimum": 1,
      "maximum": 100
    },
    "country": {
      "type": "string"
    },
    "limit": {
      "type": "integer",
      "minimum": 1,
      "maximum": 20
    }
  },
  "required": [],
  "additionalProperties": false
}
🟢get_price_index(group, since)

Chain-linked LLM price index (CN, US or all models), 30-day change, median price and series.

输入模式

{
  "type": "object",
  "properties": {
    "group": {
      "type": "string",
      "enum": [
        "CN",
        "US",
        "ALL"
      ]
    },
    "since": {
      "type": "string",
      "description": "YYYY-MM-DD"
    }
  },
  "required": [],
  "additionalProperties": false
}
🟢list_price_changes(since, min_change_pct, limit)

Official list-price cuts/increases observed in daily snapshots, and models newly added to tracking.

输入模式

{
  "type": "object",
  "properties": {
    "since": {
      "type": "string",
      "description": "YYYY-MM-DD, default last 90 days"
    },
    "min_change_pct": {
      "type": "number"
    },
    "limit": {
      "type": "integer",
      "minimum": 1,
      "maximum": 100
    }
  },
  "required": [],
  "additionalProperties": false
}
🟢estimate_self_hosting(model, gpus)

VRAM, GPUs needed, speed and hardware cost to self-host an open-weights model. Omit model to list available models.

输入模式

{
  "type": "object",
  "properties": {
    "model": {
      "type": "string"
    },
    "gpus": {
      "type": "array",
      "items": {
        "type": "string"
      },
      "description": "gpu ids, e.g. h100_sxm, h20, l20, ascend910b, rtx5090d"
    }
  },
  "required": [],
  "additionalProperties": false
}
🟢list_cloud_agents(kind, query, limit)

Cloud agents (one persistent agent per person, team agents, hosting platforms) with specs and monthly prices.

输入模式

{
  "type": "object",
  "properties": {
    "kind": {
      "type": "string",
      "enum": [
        "personal",
        "work",
        "platform"
      ]
    },
    "query": {
      "type": "string"
    },
    "limit": {
      "type": "integer",
      "minimum": 1,
      "maximum": 50
    }
  },
  "required": [],
  "additionalProperties": false
}
🟢recommend_coding_pc(category, tier, query)

Relative speed and tier (best/recommended/minimum/not_advised) of Macs, laptops and desktops for running Codex / Claude Code.

输入模式

{
  "type": "object",
  "properties": {
    "category": {
      "type": "string",
      "enum": [
        "mac",
        "laptop",
        "desktop"
      ]
    },
    "tier": {
      "type": "string",
      "enum": [
        "best",
        "recommended",
        "minimum",
        "not_advised"
      ]
    },
    "query": {
      "type": "string"
    }
  },
  "required": [],
  "additionalProperties": false
}
🟢list_ai_funding(region, sector, query, since, min_amount_usd, ...)

Funding rounds of AI companies (mainly US and China): company, business, round, amount, valuation, investors, source links. Newest first; filter by region, sector, company/investor name, date or minimum amount.

输入模式

{
  "type": "object",
  "properties": {
    "region": {
      "type": "string",
      "enum": [
        "US",
        "CN",
        "OTHER"
      ]
    },
    "sector": {
      "type": "string",
      "enum": [
        "foundation_model",
        "ai_app",
        "agent",
        "chip",
        "compute",
        "robotics",
        "autonomous",
        "infra_tools",
        "vertical",
        "other"
      ]
    },
    "query": {
      "type": "string",
      "description": "company, product or investor name"
    },
    "since": {
      "type": "string",
      "description": "YYYY-MM-DD"
    },
    "min_amount_usd": {
      "type": "number"
    },
    "sort_by": {
      "type": "string",
      "enum": [
        "date",
        "amount"
      ]
    },
    "limit": {
      "type": "integer",
      "minimum": 1,
      "maximum": 100
    }
  },
  "required": [],
  "additionalProperties": false
}

社区

评价此服务器

证据

最近观测

已验证未记录版本9 个工具