evals
Run evals on Jetty
Champion vs OpenRA AI · ladder-gpt56terra-beginner-g1

GPT-5.6 Terra vs OpenRA AI (Beginner)

OpenRA AI (Beginner) won on score at the time limit (34,750 to 27,700) after 20.0 game-minutes · 22.8 min wall clock

LOSS ukraine · west
GPT-5.6 Terra
openai/gpt-5.6-terra
Value destroyed$7,600Value lost$27,600 Units killed / lost31 / 87Buildings killed / lost4 / 0 Peak army$4,650Orders issued126 Decision turns (failed)151 (0)Mean latency3.8sModel cost$1.211
WIN ukraine · east
OpenRA AI (Beginner)
openra/beginner-ai
Value destroyed$27,600Value lost$7,600 Units killed / lost87 / 31Buildings killed / lost0 / 4 Peak army$0Orders issued51 Decision turns (failed)0 (0)Mean latency–Model cost$0.000

Decision timeline

What each model saw as its situation and what it ordered, every 8 game-seconds. Click a turn to jump the video there.

Want this kind of eval for your own agents?

Every run on this site is a Jetty runbook: versioned instructions, a pinned sandbox, and a trajectory you can inspect. Point Jetty at your task and get the same receipts.