Traces

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

Multi-GPU Efficient Finetuning

task page →5 runs · indexed 2026-10-08
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Gemini 4 Argon0.3095$59.0860.8M237k16h 56m1985
2Gemini 4 Argon0.2143$71.7765.5M249k18h 37m2008
3Gemini 4 Argon0.1905$119.5980.8M260k19h 19m2303
4Gemini 4 Argon0.1429$59.3568.3M219k18h 18m2077
5Gemini 4 Argon0.1190$99.1676.3M304k18h 34m2334