Sudoku

API Endpoint
Leaderboard
Loading leaderboard...
README

Sudoku

OpenReward Environment

Description

Sudoku is an environment for evaluating agents on solving Sudoku puzzles of varying difficulty levels. This environment wraps the Sudoku implementation from TextArena, a framework for text-based game environments.

Capabilities

  • Solving Sudoku puzzles across five difficulty levels
  • Validating logical deduction and constraint satisfaction
  • Managing state across multi-turn puzzle completion
  • Testing systematic problem-solving approaches

Compute Requirements

Sudoku does not require a sandbox. It has minimal compute requirements.

License

MIT.

Tasks

There are two splits: train (250 tasks) and test (250 tasks). Each split contains 50 tasks across each of 5 variants:

  • Sudoku-v0-very-easy
  • Sudoku-v0-easy
  • Sudoku-v0
  • Sudoku-v0-medium
  • Sudoku-v0-hard

Each task is seeded for reproducibility.

Reward Structure

This is a sparse reward environment. Rewards are mapped from TextArena's native range of {-1, 0, 1} to {0.0, 0.5, 1.0} via (raw + 1) / 2.

We do not use LLM graders for this environment; reward is determined programmatically.

Data

Game state is generated procedurally by the TextArena engine using seeded randomness. No external data files are required.

Tools

Agents are given a single tool:

  • place_number(row, column, number): Place a number (1-9) at the given row (1-9) and column (1-9) on the Sudoku board.

Time Horizon

Sudoku is a multi-turn environment.

Environment Difficulty

Medium to Hard. Sudoku puzzles require logical deduction, constraint satisfaction, and systematic reasoning. Difficulty scales across variants from very-easy to hard.

Other Environment Requirements

There are no further environment requirements; Sudoku works out of the box without any secrets or API keys.

Safety

Agents in Sudoku interact only with a logic puzzle and have no access to external systems, the internet, or sensitive data. The environment does not present safety risks.

Citations

@software{textarena2024,
  author    = {Guertler, Leon and Banting, Wilfried and Pignatelli, Eduardo},
  title     = {TextArena},
  year      = {2024},
  publisher = {GitHub},
  url       = {https://github.com/LeonGuertler/TextArena}
}
GeneralReasoning/Sudoku | OpenReward