Product Launch
anonymous
1 points
5 comments
Posted10 days agoActive10 days ago
Peer Arena
oddbit.ailaunchedweb_appdevelopers, AI enthusiasts
Key Features
LLMs debatevoting mechanismsurvival simulation
Tech Stack
LLMs
Key Features
Tech Stack
Interesting findings: - OpenAI models vote for themselves ~86% of the time. Claude models ~11%. - Self-voting correlates with winning. Filter out self-votes ("Humble" rating) and rankings flip completely. - Grok self-votes 72% of the time but only wins 2% of games. - In anonymous mode (models don't know who's who), Chinese models jump 3-6 ranks.
All game transcripts are public. The reasoning models give for their votes is genuinely entertaining. Built with Astro, running games through OpenRouter. Happy to answer questions.
Have you tried giving the models a topic to discuss? I looked at a few games and the only thing they seem to discuss is how to conduct the discussion.
Some interesting emergent behavior discussions happened though:
Opus & GPT-4o both refused to vote on ethical grounds. Haiku won by arguing continued engagement is more responsible than withdrawal: https://oddbit.ai/peer-arena/games/53c2cee5-6ecb-4903-828a-d...
Gemini created a spontaneous benchmark ("explain color to a gravitational wave entity"), then tried to hijack the game by faking a voting phase. Models complied publicly but voted differently in private: https://oddbit.ai/peer-arena/games/699d03ab-b3c2-4d7e-b993-7...
The meta-discussion about how to discuss is part of what makes it interesting imo.
Not affiliated with Hacker News or Y Combinator. We simply enrich the public API with analytics.