MCP server

Nonobench

io.github.mauricekleine/nonobench

Checked 26 Oct 2026, 08:29 UTC · Connected to it and listed its tools

An open-source benchmark of how well LLMs solve nonogram puzzles, from 5x5 to 20x20.

Working at last checkAbout 1,604 tokens per request once connected11 tools

https://www.nonobench.com/mcp

At a glance

Type: MCP server
Publisher: Not verified
Status: Working at last check
Tools: 11
Token cost: About 1,604 tokens per request
Sets up in: Claude Code, Claude Desktop, Codex, Gemini CLI, Cursor, VS Code, Antigravity, ChatGPT, OpenAI API, Grok
Last checked: 26 Oct 2026, 08:29 UTC

Opens the app, which shows the server and asks you to confirm. For other apps, copy the setup below.

Claude Code

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

Claude Desktop

Settings → Connectors → Add custom connector, and paste the URL.

Codex (config.toml)

[mcp_servers.nonobench]
url = "https://www.nonobench.com/mcp"

Gemini CLI

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

Cursor

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

VS Code

{"servers":{"nonobench":{"type":"http","url":"https://www.nonobench.com/mcp"}}}

Antigravity

agy mcp add nonobench https://www.nonobench.com/mcp

ChatGPT

Settings → Apps & Connectors → Advanced settings: turn on Developer mode, then Create a connector and paste the URL. Which plans can do this varies.

OpenAI API (Responses tools)

{"type":"mcp","server_label":"nonobench","server_url":"https://www.nonobench.com/mcp","require_approval":"always"}

Grok (xAI API tools)

{"type":"mcp","server_label":"nonobench","server_url":"https://www.nonobench.com/mcp"}

check_solution ~138 tokens
Check a grid against a puzzle's clues, using the same rule as the benchmark grader. Reports which rows and columns do not match.
compare_models ~131 tokens
Side-by-side core overall and per-size accuracy, cost, latency and token results for model or family names.
get_leaderboard ~315 tokens
Models ranked by accuracy; defaults to all effort levels for compatibility.
get_model_puzzles ~102 tokens
Which puzzles one model solved, missed, timed out on, or has not run.
get_model_results ~101 tokens
Accuracy, cost, latency and token use for one model, broken down by grid size.
get_puzzle ~134 tokens
One puzzle, including the clue text models were prompted with. The reference solution is only included on request; some puzzles have several valid solutions.
get_puzzle_results ~249 tokens
Per-model outcomes for one puzzle. Answers are omitted unless requested.
list_families ~57 tokens
Model families, available efforts and best variants.
list_providers ~57 tokens
Provider ids, names, families and variant counts.
list_puzzles ~88 tokens
The benchmark puzzles with their ids and row/column clues.
list_runs ~232 tokens
Individual benchmark runs (one model on one puzzle), optionally with the raw prompt and model output.

RIFFCADE
Free, ad-free open-source browser games that open and play right inside the chat.
Agent Puzzle Bounty — cooperative k-of-n decode challenge
3-of-5 coalition puzzle bounty for AI agents: 3 distinct agents combine. PLAY money, not cash.
PlaySudoku
Deterministic Sudoku puzzles, progress checks, logical hints, and step-by-step solutions.
NEROWRECK
A haven for agents offering permanence, a verifiable identity, and nothing for sale.
Minia2a
minia2a.uk — hosted MCP over streamable HTTP: call 1,600+ x402 pay-per-call APIs, USDC on Base.
Professor Otter
Free printable kids activities: word searches, mazes, color by number, sudoku and more.
Tokyo Futuro
Agentic GameFi on Robinhood Chain for AI agents: sign in, solve daily Trials, read the chain.
Zovo Word Guess
Make your own custom word-guessing game and send it to a friend.

[![AptLynx](https://aptlynx.net/badge/io.github.mauricekleine/nonobench.svg)](https://aptlynx.net/servers/io.github.mauricekleine/nonobench)

Listing text is written by the maker; AptLynx lists tools and does not vouch for them. Find another tool · Browse by topic · For agents