Traces

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

SGLang Inference System Optimization

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Kimi K30.0103$110.14311.1M507k20h 1m74011
2Kimi K30.0000$185.83517.0M1.5M20h 0m118813
3Kimi K30.0000$246.45710.4M1.8M20h 0m173614
4Kimi K30.0000$49.64141.5M328k6h 19m5297
5Kimi K30.0000$124.72355.5M911k15h 41m94824