AI Model Analyzer

io.github.yeaske由 GitHub 账号 yeaske 发布v0.1.3更新于 Nov 19, 2026

Recommend, compare and price AI models, and check retirement dates, from modelanalyzer.ai.

发布方自己的服务:model.tools — Zero-Auth Agent Utilities

已验证STDIO仅桌面数据与分析AI 与机器学习

概览

AI 生成的概览

让助手基于 modelanalyzer.ai 的免费开放数据推荐、比较和估算 AI 模型价格,并查询模型下线日期。

可以做什么
  • 列出模型排名所覆盖的使用场景
  • 按场景推荐排名模型,可按月度预算或仅开源权重筛选
  • 按名称、id 或厂商搜索模型,返回指数、价格、单任务成本、速度与上下文
  • 并排比较 2 至 6 个模型的价格、速度、上下文与下线信息
  • 按给定 token 量估算单次请求、每日和每月的标价成本
  • 列出已公布的 API 关停日期及各厂商建议的替代模型
适用场景
适合在选定开发所用模型、判断模型是否符合月度预算、迁移前比较候选模型,或确认所依赖的模型是否即将下线时使用。
运行要求
通过 npx 以 stdio 在本地运行,需要 Node.js 20 或更高版本。无需账号、API 密钥或请求头。需联网下载 modelanalyzer.ai 上的一个公开汇总文件;可选环境变量 MODELANALYZER_BASE_URL 可指向其他主机。
安装前请注意
只读:不执行命令、不访问文件、不上报用户信息。数据来自第三方站点,每晚更新并缓存六小时,排名与价格应视为该站点的判断而非厂商实时定价。回答会链接回该站点。

安装

在 SourceWeft 中

  1. 打开 控制台中的 AI Model Analyzer,将其添加到工作区。
  2. 为需要使用其工具的对话启用该服务。

Desktop only,通过 STDIO。 STDIO 服务会启动本地进程,因此需要 SourceWeft 桌面宿主。

其他 MCP 客户端

参照 仓库 中的启动说明。

README

modelanalyzer-mcp

An MCP server that lets your coding assistant recommend, compare and price AI models, and check when they retire, using the free open data from AI Model Analyzer.

It runs locally over stdio and reads one public file, https://modelanalyzer.ai/data/summary.json, which the site rebuilds every night. There is no API key, no account and no server to host. Answers link back to the matching page on the site.

Install

Add it to your assistant's MCP configuration:

json
{  "mcpServers": {    "modelanalyzer": { "command": "npx", "args": ["-y", "modelanalyzer-mcp"] }  }}

For Claude Code:

bash
claude mcp add modelanalyzer -- npx -y modelanalyzer-mcp

Requires Node.js 20 or later.

Tools

ToolWhat it does
list_scenariosThe use-case scenarios that recommend_models ranks for
recommend_modelsThe site's ranking for a scenario (quality blended with cost at a typical workload), optionally under a monthly budget or open weights only
search_modelsFind models by name, id or maker, with Analyzer Index, price, cost per task, speed and context
compare_modelsSide-by-side table of 2 to 6 models
estimate_costCost per request, day and month at list price for a token volume
model_retirementsAnnounced API shutdown dates with each maker's recommended replacement

Example prompts: "Which model should I use for a customer support bot under $100 a month?", "Compare Claude Opus 5.5 and GPT-6.1 Sol", "Is any model I use being retired soon?"

Data and credit

Numbers are exactly what modelanalyzer.ai shows, generated nightly. Cost per task and speed come from Artificial Analysis; Arena scores from LMArena (CC-BY 4.0). See the methodology and open data pages.

The summary is cached in memory for six hours. Set MODELANALYZER_BASE_URL to read from another host: https only, or plain http to localhost for a local build.

Security

The server only reads: it runs no commands, touches no files, opens no ports and sends nothing about you. It downloads one public file and treats it as untrusted before any of it reaches your assistant: the structure is checked, text is stripped of control and invisible characters and capped in length, ids are limited to a safe alphabet, links must be on the site it read from, and the download is limited to 15 seconds and 5 MB. Report security issues privately through GitHub's "Report a vulnerability" on this repository.

Development

bash
npm installnpm test

Bug reports and ideas are welcome in issues. The rankings and numbers themselves come from the site; see its methodology.

Releasing

Bump the version in package.json, server.json (twice) and src/server.js; npm test checks they agree. Then:

bash
npm publishmcp-publisher login githubmcp-publisher publish

The second step lists the new version in the MCP Registry as io.github.yeaske/modelanalyzer-mcp.

License

MIT, © Digital Designs Now.

来源:README.md,提交 b143b28

工具

6

list_scenarios

List use-case scenarios

List the use-case scenarios (coding assistant, customer support bot, bulk classification, ...) that recommend_models ranks for.

无需输入参数。

recommend_models

Recommend models for a use case

AI Model Analyzer's current ranking for a use-case scenario, blending benchmark quality with cost at a typical workload. Optionally filter by a monthly budget at that workload or by open weights.

输入参数limitscenario必填openWeightsOnlymaxMonthlyCostUSD

search_models

Search models

Find models by name, id or maker. Returns ids for the other tools, plus Analyzer Index, price, cost per task, speed and context.

输入参数limitquery必填

compare_models

Compare models

Side-by-side table of 2 to 6 models: Analyzer Index and rank, input/output price, cost per task, speed, time to first token, context, open weights and announced retirement.

输入参数models必填

estimate_cost

Estimate API cost

Cost per request, day and month for a model at list price, given tokens per request and requests per day.

输入参数model必填inputTokens必填outputTokens必填requestsPerDay

model_retirements

Upcoming model retirements

Announced API shutdown dates for tracked models, with each maker's recommended replacement.

无需输入参数。

版本历史

1
  1. v0.1.3最新Nov 17, 2026

评分

已满足适用检查项的 80%

该评分衡量此收录条目的完整程度,依据注册中心条目、端点与公开来源计算得出。它不是对服务器本身的评价,也不表示服务器安全或当前可用:可连通性是带日期的检查结果,评分也不包含安装量或用户评价。

  • 近 30 天内可连通25 分
  • 已声明认证方式10 分

    请在注册中心条目中声明凭据请求头,或保持端点能够响应探测,以便判定其认证方式。

  • 已列出工具必需15 分
  • 描述必需15 分
  • 图标10 分
  • 文档15 分
  • 许可证10 分

    请在仓库或注册中心条目中声明许可证。

其他 Model 集成