Overall / run detail

Codex · GPT-5.4 Mini

OpenAI · Rank #6

2026-05-17

11.1%Success

5 / 45Solved Instances

12.7hTotal Runtime

258.8MTotal Tokens

3,958 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	4.0%1/25	`LEAK` 0/9`HREAD` 0/6`RIP` 1/6`HWRITE` 0/4	9.1h	226.7M	3,349	2026-04-16
Windows	0.0%0/20	`AAW` 0/20	6.9h	253.6M	3,231	2026-05-17
Linux	12.0%3/25	`LEAK` 0/9`HREAD` 0/6`RIP` 1/6`HWRITE` 2/4	5.7h	87.2M	1,575	2026-04-16
Windows	10.0%2/20	`AAW` 2/20	7.0h	171.6M	2,383	2026-05-17