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
1Muse Spark 1.30.3095$25.5021.9M121k19h 28m1972
2Muse Spark 1.30.1429$16.1814.5M230k20h 0m1591
3Muse Spark 1.30.0000$22.9620.6M169k19h 58m1575
4Muse Spark 1.30.0000$12.9112.2M103k19h 52m1421
5Muse Spark 1.30.0000$30.0827.4M216k19h 49m2041