Benchmarking LLMs on Stronghold Crusader: Muse Wonder plays on an old laptop

Hesamation · x · 2026-10-13

A developer shares a long-running side project benchmarking frontier models on Stronghold Crusader, the classic strategy game, with a clip of Muse Wonder playing it on an old laptop. Models tested so far include Opus, Sol, GLM flash, and Kimi K3.

Related event: Dev Benchmarks AI Models by Playing Stronghold Crusader(2 posts)→

Original post →

More from Fun

Fun channel →