Skip to content
MCP server

AI Compute Radar

Open models ranked by momentum, with GPU or Mac fit verdicts and GPU rental prices

Install
Pricing
Free
Platforms
API, CLI, Web
Runs
Locally or Remotely
Version
1.2.3
Listing updated
2 days ago

About AI Compute Radar

AI Compute Radar is a free MCP server for ai models and llm tooling. It can run locally on your machine or as a hosted remote server.

AI Compute Radar publishes measured figures rather than benchmarks. Its API and MCP server return no speed numbers, benchmark scores or forecasts, and where a reading is missing the field is null instead of a guess. Every response carries a meta block saying whether data is live or a demo fixture, when it was collected and how to cite it.

The hardware-fit tool compares a model's measured weights from its GGUF file listing, a context cache computed from the model's config, and a declared overhead assumption against usable memory on a chosen GPU or Mac, at a chosen context length and KV cache type, and gives plain-language reasons. The momentum board ranks tracked models with a Heat Score built from Hugging Face and OpenRouter signals. Price tools report median Vast.ai on-demand rental per GPU class, alongside RunPod, Clore.ai and AWS, Azure and Oracle list prices, with up to 365 days of history from 26 August 2026. A weekly pick comes from a published rule.

On 13 November 2026 the API served 49 models, 37 hardware profiles and rental prices for 8 GPU classes from 11 sources. Data is CC BY 4.0, and the server needs no key; a stdio npm bridge is also provided.

Install

Run in a terminal

claude mcp add --transport http ai-compute-radar https://aicomputeradar.dev/api/mcp

Add to Settings → Connectors → Add custom connector

https://aicomputeradar.dev/api/mcp
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "ai-compute-radar": {
      "url": "https://aicomputeradar.dev/api/mcp"
    }
  }
}
Install in VS CodeOpens VS Code, which asks you to confirm. Or add it by hand:

Add to .vscode/mcp.json

{
  "servers": {
    "ai-compute-radar": {
      "type": "http",
      "url": "https://aicomputeradar.dev/api/mcp"
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "ai-compute-radar": {
      "serverUrl": "https://aicomputeradar.dev/api/mcp"
    }
  }
}

Add to your client's MCP settings (most use this shape)

{
  "mcpServers": {
    "ai-compute-radar": {
      "type": "http",
      "url": "https://aicomputeradar.dev/api/mcp"
    }
  }
}

Tools this server provides (5)

Taken from the project documentation; the running server may differ.

trending_models
Tracked models ranked by Heat Score with the signals behind each.
find_fit
Fit verdicts for a model on a chosen GPU or Mac and context length.
gpu_prices
Median GPU rental prices per class and source.
list_hardware
Lists curated GPU and Mac profiles.
weekly_pick
The model chosen by the published weekly rule.

What it does

Hardware-fit verdicts
Checks whether a model fits a chosen GPU or Mac at a given context length.
Model momentum board
Ranks open models by a Heat Score from Hugging Face and OpenRouter signals.
GPU rental prices
Median on-demand prices per GPU class with offer counts and list-price comparisons.
Price history
Daily history per class and source up to 365 days, as JSON or CSV.
Provenance in responses
Each answer says whether data is measured or fixture and when it was collected.
Weekly pick
A model chosen by a published rule with its report and caveats.

Good for

  • Check whether a model fits an RTX 4090 at 16K context
  • Compare GPU rental prices across providers
  • Spot which open models are trending
  • Pull rental price history into a spreadsheet
  • Cite measured data with attribution in a report

Pricing

Free

Free; data under CC BY 4.0

Prices change. Check aicomputeradar.dev before you buy.

AI Compute Radar FAQ

Do I need an API key?

No. The MCP server and read-only JSON API need no keys.

What licence is the data under?

CC BY 4.0, with attribution included in every response.

Does it give benchmarks or speed figures?

No. The site says there are no benchmark results, speed figures or forecasts in any field.

Which tools does the server offer?

trending_models, find_fit, gpu_prices, list_hardware and weekly_pick.

How do stdio-only clients connect?

Run npx -y ai-compute-radar, a dependency-free bridge that holds no data and sends no telemetry.

Is AI Compute Radar free?

Yes. AI Compute Radar is free to use.

How do I add AI Compute Radar to Claude Code?

Run this command in your terminal: claude mcp add --transport http ai-compute-radar https://aicomputeradar.dev/api/mcp

What are the best alternatives to AI Compute Radar?

The closest alternatives to AI Compute Radar are Agent-Native Analytics, Claude Code Ultimate Guide, Google Ads + Meta Ads + SEO/GEO MCP, Tapetide - Indian Stock Market MCP and World Monitor.

Alternatives to AI Compute Radar

  1. Open-source analytics app where an AI agent builds charts and dashboards from your data

    • Remote
    • Open source
    • 7.7K GitHub stars
    • Analytics and data
  2. MCP server for searching the Claude Code Ultimate Guide from any MCP client

    Runs locally
    • Open source
    • 6.5K GitHub stars
    • AI models and LLM tooling
  3. Hosted MCP connector for managing Google Ads, Meta Ads, GA4 and Search Console from chat

    • 8 tools
    • Remote
    • Freemium
    • 8.1K GitHub stars
    • Marketing and SEO
  4. Indian NSE and BSE stock market data for AI assistants through 48 MCP tools

    • 17 tools
    • Remote
    • Freemium
    • 150 GitHub stars
    • Finance and payments
  5. World Monitor

    MCP server

    Live geopolitical intelligence dashboard with an API and MCP server for agents

    • Remote
    • Freemium
    • Analytics and data
  6. CompanyData

    MCP server

    Search and enrich company records from official registers via the CompanyData API

    • 3 tools
    • Remote
    • Paid
    • 3 GitHub stars
    • Analytics and data