Traces

Every run: the agent's full trajectory, its submissions over the 20-hour budget, and the final score.

Snooker Prediction

task page →5 runs · indexed 2026-09-22
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Gemini 3.8 Flash0.1105$14.99133.6M331k1h 42m4033
2Gemini 3.8 Flash0.0820$13.08117.4M334k1h 20m3683
3Gemini 3.8 Flash0.0000$8.1176.9M253k1h 6m2995
4Gemini 3.8 Flash0.0000$8.7083.2M282k1h 15m3121
5Gemini 3.8 Flash0.0000$17.30177.6M409k1h 51m4715