evals
Run evals on Jetty
Round robin · rr-jev-gpt56terra-g1

Jev Router vs GPT-5.6 Terra

GPT-5.6 Terra destroyed the enemy base after 5.7 game-minutes · 8.1 min wall clock

LOSS russia · west
Jev Router
typesafe/jev-router
Value destroyed$3,600Value lost$8,850 Units killed / lost34 / 23Buildings killed / lost0 / 5 Peak army$1,950Orders issued79 Decision turns (failed)44 (0)Mean latency4.8sModel cost$0.067
WIN russia · east
GPT-5.6 Terra
openai/gpt-5.6-terra
Value destroyed$8,850Value lost$3,600 Units killed / lost23 / 34Buildings killed / lost5 / 0 Peak army$5,000Orders issued91 Decision turns (failed)44 (0)Mean latency3.6sModel cost$0.323

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.