GermanWhist

Description

GermanWhist is an ORS environment for evaluating agents on trick-taking card game strategy. This environment wraps the GermanWhist implementation from TextArena, a framework for text-based game environments.

Capabilities

• Trick-taking game mechanics and strategy
• Suit-following rules and card play optimization
• Competitive gameplay against an LLM opponent
• Sequential decision-making with partial information

Compute Requirements

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

License

MIT.

Tasks

There are two splits: train (150 tasks) and test (150 tasks). Each split contains 50 tasks across each of 3 variants:

• GermanWhist-v0
• GermanWhist-v0-train
• GermanWhist-v0-raw

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:

• play_card(card_number): Play a card from your hand by number.

Time Horizon

GermanWhist is a multi-turn environment.

Environment Difficulty

Medium - requires understanding trick-taking mechanics, suit-following rules, and strategic card selection to maximize tricks won.

Other Environment Requirements

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

Safety

Agents in GermanWhist interact only with a card game simulation 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}
}