Traces

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

Reconnaissance Blind Chess Recovery

task page →5 runs · indexed 2026-09-02

#	Model	Reward	Cost	In tok	Out tok	Time	Steps	Submits
1	Kimi K3	0.7571	$29.23	77.1M	277k	14h 20m	314	2
2	Kimi K3	0.5071	$31.51	82.4M	325k	14h 46m	311	4
3	Kimi K3	0.4286	$27.43	64.3M	286k	14h 50m	256	11
4	Kimi K3	0.3143	$28.91	69.9M	341k	12h 50m	253	2
5	Kimi K3	0.1429	$43.61	116.1M	397k	14h 40m	395	2