Nonobench

Connect it to your AI

Claude Code

claude mcp add --transport http nonobench https://www.nonobench.com/mcp

Claude, ChatGPT, Cursor and other apps

Add it as a custom connector with this URL (Streamable HTTP):

https://www.nonobench.com/mcp

Cursor, VS Code and other apps (mcp.json)

{
  "mcpServers": {
    "nonobench": {
      "url": "https://www.nonobench.com/mcp"
    }
  }
}

Tools it offers

11 tools, as the server itself reported during our check on Oct 31, 2026.

• get_leaderboardModels ranked by accuracy; defaults to all effort levels for compatibility.
• list_providersProvider ids, names, families and variant counts.
• list_familiesModel families, available efforts and best variants.
• compare_modelsSide-by-side core overall and per-size accuracy, cost, latency and token results for model or family names.
• get_model_resultsAccuracy, cost, latency and token use for one model, broken down by grid size.
• list_puzzlesThe benchmark puzzles with their ids and row/column clues.
• get_puzzleOne puzzle, including the clue text models were prompted with. The reference solution is only included on request; some puzzles have several valid solutions.
• check_solutionCheck a grid against a puzzle's clues, using the same rule as the benchmark grader. Reports which rows and columns do not match.
• get_puzzle_resultsPer-model outcomes for one puzzle. Answers are omitted unless requested.
• get_model_puzzlesWhich puzzles one model solved, missed, timed out on, or has not run.
• list_runsIndividual benchmark runs (one model on one puzzle), optionally with the raw prompt and model output.

Questions