MCP serverCommunity

evalpair

Run LLM eval suites on local or hosted models; compare runs with intervals and a noise floor.

Srikarmk · ↓ 0 installs

Install

Add the config to your client settings. Replace values in <angle brackets> with your own.

claude_desktop_config.json

{
  "mcpServers": {
    "evalpair": {
      "command": "uvx",
      "args": [
        "evalpair==0.1.1"
      ],
      "env": {
        "EVALPAIR_HOME": "<EVALPAIR_HOME>",
        "EVALPAIR_API_KEY": "<EVALPAIR_API_KEY>"
      }
    }
  }
}

Or in one command — the CLI adds the server to your client config:

CLI

npx ai-point add io.github.Srikarmk/evalpair

Tools

The tool list will appear once the server has a sandbox or the editors fill it in.

Report a problem ↗ (Opens GitHub in a new tab)