ai-market-intelligence
AI Market Intelligence MCP
一个用 TypeScript 编写的本地 MCP 服务器,用于按场景比较 AI 模型:编程、支持、文档、多模态、代理和隐私。
此第一版的功能
列出本地模型目录;
根据能力、上下文、相对成本和自托管的明确标准推荐模型;
比较选定的模型;
解释每个场景中使用的权重。
src/catalog.ts 中的配置文件是定性的初步文件:不是官方的价格或基准。在做出购买决定之前,请使用您所在领域的评估来验证推荐,并更新来源。
Related MCP server: LLM Benchmark MCP Server
运行
npm install
npm run build
npm run dev要使用 MCP Inspector 进行测试:
npm run inspect在 VS Code 中连接
在将使用该服务器的项目中创建或更新 .vscode/mcp.json:
{
"servers": {
"ai-market-intelligence": {
"command": "node",
"args": ["C:/Users/gabiv/Projects/ai-market-intelligence-mcp/dist/index.js"]
}
}
}运行 npm run build 后,打开 VS Code/Copilot 的 MCP 部分并启动服务器。传输方式是 stdio;因此,该程序从不在 stdout 中写入日志。
建议的后续演进
用于数据和版本历史的 JSON 文件或 SQLite 数据库。
官方来源收集器(价格、上下文窗口和限制),包含每个数据的日期和 URL。
使用真实提示词和您的业务指标进行评估,而不是依赖通用的比较。
带身份验证的 HTTP 界面,如果您想与团队共享服务器。
Available Tools
4 toolscompare_modelsComparar modelosB
Compara até quatro modelos para um cenário específico e mostra a classificação.
Inputs: ids (required), scenario (required).
explain_scenariosExplicar cenáriosA
Explica os cenários avaliados e os pesos usados pelo mecanismo de recomendação.
Inputs: no named parameters.
list_modelsListar modelosB
Lista os perfis de modelos presentes no catálogo local.
Inputs: provider.
recommend_modelRecomendar modeloC
Ordena modelos do catálogo para um cenário, orçamento e requisito de hospedagem própria.
Inputs: scenario (required), limit, budget, requiresSelfHosted.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 4 tools
The tools are mostly distinct: list_models lists catalog items, recommend_model ranks them, compare_models compares a subset, and explain_scenarios describes the reasoning. However, recommend_model and compare_models could be confused since both rank/compare for a scenario, though compare_models explicitly limits to four.
All tools follow a verb_noun pattern (list_models, recommend_model, compare_models, explain_scenarios). However, the verbs vary in style - 'list', 'recommend', 'compare' are action verbs, while 'explain' is also action but the noun 'scenarios' differs from 'models' - but the pattern is consistent.
With 4 tools, the set is slightly small for a domain that includes model cataloging, recommendation, comparison, and explanation, but it is reasonable for a focused utility. Each tool serves a distinct function and the count is acceptable, though a bit sparse.
The domain appears to be model recommendation and comparison, and the tools cover listing, recommending, comparing, and explaining the logic. However, there is no tool to get details of a single model or to manage the catalog (add/update/delete), which are notable gaps for a complete surface.
Maintenance
Related MCP Connectors
Live prices, rankings, licences and privacy of AI models: best model per task, costs, alternatives.
Which LLM should I use for this task? Ask in plain words and get the models that score best on the matching benchmark board, within your price, context window and open-weights limits. Covers 400+ models from OpenAI, Anthropic, Google, DeepSeek, Qwen, Mistral and more, with benchmark scores (coding, agentic, reasoning, math, knowledge, design) and API list prices per million tokens, updated daily. Tools: recommend_model, get_leaderboard, compare_models, get_model, get_prices, list_categories. Free, no API key, data under CC BY 4.0. Docs: https://llmbenchmark.ai/api
Find open-weight AI models that fit your hardware, with weight sizes, memory and sources.
Find the right AI model for your job. Free discovery and ranking; tests on your own key.
Related MCP Servers
- AlicenseDqualityDmaintenanceDescribe your AI use case in plain English, get ranked model recommendations with cost estimates and tradeoff reasoning. Covers 62 models across 29 providers. Available as a web app (BYOK + guest tier) and as an MCP server for Claude Desktop and Cursor — same recommendation engine, two interfaces.11MIT
- AlicenseNot gradedqualityDmaintenanceEnables AI agents to compare LLM benchmarks, get pricing, and receive model recommendations for tasks and budgets.1MIT
- FlicenseNot gradedqualityDmaintenanceEnables comparing outputs of 17 large language models side by side, with AI judging, cost and speed analysis.-
- AlicenseAqualityCmaintenanceEnables users to select the optimal LLM for their specific task by aggregating benchmark data and user intent, returning a ranked shortlist with plain-English rationale.5MIT