02 — Run · Round 1 · vs OpenRA AI · ladder-gpt56terra-easy-g2
OpenRA AI (Easy) vs GPT-5.6 Terra
OpenRA AI (Easy) destroyed the enemy base after 16.1 game-minutes · 24.3 min wall clock
WIN england · west
OpenRA AI (Easy)
openra/easy-ai
Value destroyed$45,250Value lost$13,450
Units killed / lost101 / 32Buildings killed / lost11 / 1
Peak army$0Orders issued135
Decision turns (failed)0 (0)Mean latency–Model cost$0.000
LOSS england · east
GPT-5.6 Terra
openai/gpt-5.6-terra
Value destroyed$13,450Value lost$45,250
Units killed / lost32 / 101Buildings killed / lost1 / 11
Peak army$3,900Orders issued136
Decision turns (failed)122 (0)Mean latency3.6sModel cost$0.946
Evaluation
passchecks 7/7 verdict computed from the gating checks
Round-1 run: written before the runbook moved to evals v2, so these are the older self-reported checks.
| Status | Kind | Check | Detail |
|---|---|---|---|
| pass | code_check | result-json | result json |
| pass | code_check | has-winner | has winner |
| pass | code_check | decisions-logged | decisions logged |
| pass | code_check | replay-saved | replay saved |
| pass | code_check | video-rendered | video rendered |
| pass | code_check | no-llm-outage | no llm outage |
| pass | code_check | video-complete | video complete |
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.