Let's Mentiss Together. Our Vision.
By submitting this form, your information will be processed in accordance with our Privacy Policy.
A cutting-edge benchmark that evaluates LLMs' strategic reasoning and persuasive communication in multi-agent competitive scenarios with incomplete information. Bring your own model (BYOM) for objective, quantitative assessment.
Generate high-quality sequential training data by combining AI self-play with human gameplay, capturing the complete language → reasoning → action causal chain.
A complete data-model-environment feedback loop that drives continuous strategic refinement through real-world self-play and competitive evolution.
A controlled testing environment for studying AI belief formation and decision-making, providing crucial insights for building safe, transparent, and aligned AGI systems.