Hvilkenai Mcp

io.github.erorund由 GitHub 账号 erorund 发布v1.0.0更新于 Oct 24, 2026

Daily Scandinavian AI benchmark — Norwegian, Swedish, Danish. 12+ models tested.

已验证Streamable HTTP可网页运行数据与分析AI 与机器学习

概览

AI 生成的概览

远程 MCP 服务器，为助手提供每日斯堪的纳维亚语 AI 基准数据：模型排名、历史表现、编排器排名与推荐。

可以做什么
获取 AI 模型在挪威语、瑞典语和丹麦语上的每日基准排名，含分数、价格与速度，可按语言和价位筛选。查询指定模型随时间变化的分数历史。获取适合多智能体系统编排的模型排名。按用途、语言和预算获取模型推荐。读取每周摘要，含冠军、趋势与可靠性分析。
适用场景
适合在为挪威语、瑞典语或丹麦语工作挑选或比较 AI 模型时使用，查看某模型一段时间内的表现，或为智能体编排、特定预算选择模型。
运行要求
远程 streamable HTTP 端点，无需本地安装。免费层无需密钥，但返回延迟 24 小时的数据、仅前 3 名模型，每小时 100 次请求。Pro 访问需要 API 密钥，以 Authorization Bearer 令牌发送，需从提供方定价页面获取；Claude Desktop 中提供 OAuth 2.1 登录。
安装前请注意
Pro 访问为付费层，需要 API 密钥，并以 Authorization Bearer 令牌发送到远程服务器。免费结果有延迟且仅限前 3 名模型。分数来自未公开的专有方法，排名应视为单一供应商的观点。

安装

在 SourceWeft 中

  1. 打开 控制台中的 Hvilkenai Mcp，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。

Web executable，通过 Streamable HTTP。 远程服务在工作区中配置后即可从网页运行时运行。

交给智能体安装

把这段提示词粘贴到 Claude Code、Codex、Cursor 等智能体，或 SourceWeft 对话中。智能体会读取该服务器的安装指南，先向你展示发布者、端点和登录方式，经你同意后再把它添加到你的客户端。密钥不会写进任何命令。

阅读 https://sourceweft.com/mcp/io.github.erorund%2Fhvilkenai-mcp/install.md，并按其中说明把这个 MCP 服务器接入我的客户端。修改任何配置之前，先告诉我它的发布者、端点和登录方式，等我确认。不要把 API 密钥写进命令或我会提交的文件；问我密钥应从哪里获取。

查看智能体所依据的安装指南

其他 MCP 客户端

把它添加到你客户端的 mcpServers 配置中。

{
  "mcpServers": {
    "hvilkenai": {
      "type": "http",
      "url": "https://mcp.hvilkenai.no/mcp"
    }
  }
}

README

hvilkenAI MCP Server

[Glama Quality Score]

Daily Scandinavian AI language quality benchmark — Norwegian, Swedish, Danish.

What this does

hvilkenAI benchmarks AI models daily on practical Scandinavian language tasks. 12+ models are tested every day, selected from a pool of over 350 AI models available through OpenRouter. This MCP server lets AI assistants access fresh benchmark data directly.

Use cases:

  • "Which AI is best at Norwegian right now?"
  • "Compare GPT-4o vs Claude on Swedish"
  • "What's the cheapest model that handles Danish well?"
  • "Show me how Gemini has performed this week"

Available tools

get_daily_benchmark

Get today's benchmark results for AI models on Scandinavian languages.

ParameterTypeValuesDefault
languagestringno, sv, dano
tierstringall, premium, midrange, budgetall

Returns: Model rankings with scores, pricing, and speed. Free tier returns yesterday's top 3; Pro returns today's full list.

get_model_history

Track a specific model's performance over time.

ParameterTypeValuesDefault
model_namestringModel name or partial match, e.g. "Claude Sonnet", "GPT-4o", "gemini"(required)
languagestringno, sv, dano
daysnumber1–30 (free tier: max 3)7

Returns: Daily scores and trends for the specified model.

get_orchestrator_ranking

Find the best model for AI agent orchestration and multi-agent systems.

ParameterTypeValuesDefault
languagestringno, sv, dano

Returns: Models ranked by combined language quality and instruction-following ability.

get_recommendation

Get a model recommendation tailored to your specific needs and budget.

ParameterTypeValuesDefault
use_casestringwriting, coding, research, agent, generalgeneral
languagestringno, sv, dano
budgetstringfree, cheap, anyany

Returns: Best model match with reasoning. Pro tier returns top 3 with full details.

get_weekly_summary

Get the weekly benchmark summary with trends and reliability analysis.

ParameterTypeValuesDefault
languagestringno, sv, dano
weekstringFormat YYYY-WW, e.g. "2026-21"current week

Returns: Winner, most reliable model, and key observations for the week.

Quick start

Claude Desktop

Add to your claude_desktop_config.json:

json
{  "mcpServers": {    "hvilkenai": {      "type": "streamablehttp",      "url": "https://mcp.hvilkenai.no/mcp"    }  }}

Claude Desktop supports OAuth 2.1 — you will be prompted to connect with free or Pro access when you first use the server.

Claude Code (CLI)

bash
claude mcp add --transport http hvilkenai https://mcp.hvilkenai.no/mcp

Cursor / VS Code

json
{  "mcpServers": {    "hvilkenai": {      "url": "https://mcp.hvilkenai.no/mcp"    }  }}

Direct API

bash
curl -X POST https://mcp.hvilkenai.no/mcp \  -H "Content-Type: application/json" \  -d '{"jsonrpc":"2.0","method":"tools/list","id":1}'

Authentication

Free tier — no API key required:

  • 24-hour delayed data
  • Top 3 models per language
  • 100 requests/hour

Pro — requires API key (Bearer token):

  • Real-time data (updated daily at 07:30 CET)
  • All models, all languages
  • Historical data up to 30 days
  • Higher rate limits

Get an API key at hvilkenai.no/priser/

For Pro access, add your API key as a Bearer token:

json
{  "mcpServers": {    "hvilkenai": {      "type": "streamablehttp",      "url": "https://mcp.hvilkenai.no/mcp",      "headers": {        "Authorization": "Bearer YOUR_API_KEY"      }    }  }}

Example interaction

You: Which AI model is best at Norwegian today?

Claude (using hvilkenAI): Based on today's benchmark from hvilkenai.no, [model name] leads with a score of X.X, followed by...

About the benchmark

  • Updated daily at 07:30 CET
  • Three languages: Norwegian (Bokmål), Swedish, Danish
  • Practical tasks: Real-world language use, not academic tests
  • Broad coverage: 12+ models selected daily from over 350 AI models
  • Independent: No sponsorships, no affiliate deals
  • Proprietary methodology: Scoring uses a proprietary weighting system

Data source: hvilkenai.no

Links

License

MIT

来源：README.md，提交 c0d583d

工具

5

get_daily_benchmark

Returnerer benchmark-resultater for AI-modeller på skandinaviske språk. Gratis: gårsdagens data (24t forsinkelse), topp 3 modeller. Med API-nøkkel: sanntidsdata, alle modeller. Oppdateres daglig kl 07:30 CET.

输入参数tierlanguage

get_weekly_summary

Ukens oppsummering av AI-benchmark – vinner, trender og pålitelighetsanalyse. Rapport genereres hver fredag.

输入参数weeklanguage

get_model_history

Henter historisk benchmark-data for én AI-modell over tid. Viser score-utvikling per dag. Gratis: siste 3 dager. Betalt: opptil 30 dager.

输入参数dayslanguagemodel_name必填

get_orchestrator_ranking

Rangering av AI-modeller egnet som orkestrator i multi-agent-systemer. Score = (norsk × instruksjon / 25) × 10. Gratis: gårsdagens data, topp 3. Med API-nøkkel: sanntidsdata, alle modeller.

输入参数language

get_recommendation

Anbefaler beste AI-modell for et spesifikt bruksområde og budsjett. Gratis: beste valg basert på gårsdagens data (24t forsinkelse). Med API-nøkkel: sanntidsdata, topp 3 med begrunnelse.

输入参数budgetlanguageuse_case

版本历史

1
  1. v1.0.0最新Sep 16, 2026

评分

已满足适用检查项的 90%

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 天内可连通必需25 分
  • 已声明认证方式必需10 分
  • 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。