Skip to content
MCP server

Stalemodel

Catch outdated AI model IDs in code and look up current LLM prices and limits

Runs locallyStatus and history
Install
Pricing
Open source
Platforms
CLI, Linux, macOS
Runs
Locally
Version
0.1.1
Listing updated
today

Similar MCP servers

About Stalemodel

Stalemodel is an open-source MCP server for ai models and llm tooling. It runs locally on your machine and is started by your AI client.

Coding agents often write model IDs from old training data, which leads to errors such as model not found. This tool checks IDs against a live catalog, suggests the newest model in the same family, and reports prices per million tokens and context sizes. It covers Anthropic, OpenAI, Google Gemini, xAI, DeepSeek and Mistral.

It runs as an MCP server over stdio and as a CLI, needs no API key or account, sends no code anywhere and falls back to a bundled snapshot offline. The scan command exits with status 1 on findings, so it can run in CI. Data comes from models.dev, so it can lag providers, and scan only sees quoted IDs.

The README says it needs Node 22.19 or newer, is MIT licensed, was largely generated with Claude Code and has so far only been tested in Claude Code.

Install

Run in a terminal

claude mcp add stalemodel -- npx -y stalemodel

Add to claude_desktop_config.json

{
  "mcpServers": {
    "stalemodel": {
      "command": "npx",
      "args": [
        "-y",
        "stalemodel"
      ]
    }
  }
}
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "stalemodel": {
      "command": "npx",
      "args": [
        "-y",
        "stalemodel"
      ]
    }
  }
}
Install in VS CodeOpens VS Code, which asks you to confirm. Or add it by hand:

Add to .vscode/mcp.json

{
  "servers": {
    "stalemodel": {
      "type": "stdio",
      "command": "npx",
      "args": [
        "-y",
        "stalemodel"
      ]
    }
  }
}

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

{
  "mcpServers": {
    "stalemodel": {
      "command": "npx",
      "args": [
        "-y",
        "stalemodel"
      ]
    }
  }
}

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

{
  "mcpServers": {
    "stalemodel": {
      "command": "npx",
      "args": [
        "-y",
        "stalemodel"
      ]
    }
  }
}

Using several servers? Build one config file with the MCP config generator.

Tools this server provides (3)

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

models_check
Says whether IDs are current, deprecated or unknown and suggests a newer model.
models_find
Lists models by provider, name, price, context size or tool support.
models_scan
Finds outdated IDs in a file or folder, returning locations only.

What it does

Deprecated ID scan
Finds outdated model names in a file or folder.
Model lookup
Lists models by provider, price, context size or tool support.
ID check
Reports whether IDs are current, deprecated or unknown.
Offline snapshot
Uses a bundled catalog without network access.
CI exit code
scan exits 1 on findings.
Claude Code plugin
Installs as a plugin with a skill.

Good for

  • Fix 'model not found' errors from stale IDs
  • Gate pull requests on deprecated model names
  • Find the cheapest current model with a 1M context
  • Pick the right Claude ID and see its cost

Pricing

Open source

MIT licensed

Stalemodel FAQ

Does it need an API key?

No key or account is needed.

Where does pricing come from?

From models.dev, fetched live with an offline snapshot.

What are its limits?

Data can lag, and scan misses IDs built at runtime.

Which providers?

Anthropic, OpenAI, Google, xAI, DeepSeek and Mistral.

Is Stalemodel free?

Yes. Stalemodel is open source, so you can use and self-host it for free.

How do I add Stalemodel to Claude Code?

Run this command in your terminal: claude mcp add stalemodel -- npx -y stalemodel

What are the best alternatives to Stalemodel?

The closest alternatives to Stalemodel are REA, Context7, Chrome DevTools MCP, Codebase Memory and GitHub.

Where this comes from

  • Description: written from the project's README.
  • Install steps: from the official MCP Registry entry.

How AIPoweredBot checks servers