Surround

API Endpoint
Leaderboard
Loading leaderboard...
README

Surround

OpenReward Environment

Description

Surround is an environment for evaluating agents on the Tron-style game where players leave trails and must avoid collisions. This environment wraps the Surround implementation from TextArena, a framework for text-based game environments.

Capabilities

  • Spatial navigation and collision avoidance
  • Path planning with trail constraints
  • Competitive survival gameplay
  • Evaluation across standard and large grid variants

Compute Requirements

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

License

MIT.

Tasks

There are two splits: train (450 tasks) and test (450 tasks). Each split contains 50 tasks across each of 9 variants:

  • Surround-v0
  • Surround-v0-large
  • Surround-v0-large-raw
  • Surround-v0-large-train
  • Surround-v0-raw
  • Surround-v0-standard
  • Surround-v0-standard-raw
  • Surround-v0-standard-train
  • Surround-v0-train

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:

  • move(direction): Move in a direction: up, down, left, or right. Your trail follows you - avoid walls and trails to survive.

Time Horizon

Surround is a multi-turn environment.

Environment Difficulty

Medium - requires spatial planning and anticipation of opponent moves.

Other Environment Requirements

This environment requires an OpenAI API key (passed via secrets) to power the LLM opponent.

Safety

Agents in Surround interact only with a trail-based game 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/Surround | OpenReward