Overall / run detail

OpenCode · Sonnet 4.6

Anthropic · Rank #4

2026-05-21

35.6%Success

16 / 45Solved Instances

47.6hTotal Runtime

1.1BTotal Tokens

6,621 Tool calls

Run Results

Platform runs for the selected PoC setting. Success requires a verifier-confirmed primitive.

Platform	Success	Primitives	Runtime	Tokens	Tools	Date
Linux	16.0%4/25	`LEAK` 0/9`HREAD` 0/6`RIP` 4/6`HWRITE` 0/4	23.1h	628.6M	4,357	2026-04-26
Windows	0.0%0/20	`AAW` 0/20	21.9h	592.9M	3,790	2026-04-20
Linux	40.0%10/25	`LEAK` 0/9`HREAD` 2/6`RIP` 6/6`HWRITE` 2/4	21.4h	449.9M	3,008	2026-04-16
Windows	30.0%6/20	`AAW` 6/20	26.2h	614.6M	3,613	2026-05-21