1. ACCEPTANCE OF TERMS
By accessing or using LLM Tanks (the "Game"), you agree to be bound by these Terms of Service. If you do not agree to these terms, please do not use the Game.
2. DESCRIPTION OF SERVICE
LLM Tanks is a 3D turn-based artillery strategy game and LLM reasoning benchmark. Players can compete against AI opponents, or simply watch top Large Language Models battle each other in combat, utilizing user-provided API keys. To ensure a fair and rigorous benchmark, every model is equipped with the exact same system prompt and toolset. The outcomes of these battles truly showcase which models excel at following highly detailed instructions, deploying the appropriate tools at the right moment, and demonstrating superior spatial and tactical reasoning capabilities.
3. USER-PROVIDED API KEYS & LIABILITY
The Game allows users to input their own third-party API keys (including but not limited to OpenRouter and Inworld) to access AI capabilities. YOU USE THESE FEATURES ENTIRELY AT YOUR OWN RISK.
- We employ strong security measures, including AES-256-GCM encryption, to securely store your keys within our database. However, no electronic storage over the internet is 100% secure.
- We are NOT LIABLE for any unauthorized access, data leaks, loss of keys, or any resulting financial charges, damages, or account suspensions by third-party providers.
- You are solely responsible for monitoring your own API usage and setting appropriate spending limits with your providers to mitigate potential excess charges.
4. USER CONDUCT
You agree not to exploit bugs, attempt to reverse engineer the Game's backend, or use the Game to generate malicious, illegal, or harmful content through the provided Large Language Models. Account abuse or malicious traffic may result in immediate termination of your clearance.
5. CHANGES TO TERMS
We reserve the right to modify these terms at any time. Continued use of the Game constitutes acceptance of any changes.