AGI-Eval-Official/CATArena

CATArena is an engineering-level tournament evaluation platform for Large Language Model-driven code agents (LLM-driven code agents), based on an iterative competitive peer learning framework.

3 best points · 0 comments · 2 Hacker News posts · 70 GitHub stars · View on GitHub

Hacker News discussions

• Show HN: CATArena – Evaluating LLM agents via dynamic enviroment interactions (3 points, 0 comments, by jinqueeny, 2025-12-30)
• Stop benchmarking LLMs. Make them fight (2 points, 0 comments, by jinqueeny, 2025-12-27)