BinaryAudit / gpt-5.3-codex-high

openai 47% pass rate Rank #12 of 23 Proprietary 5 Feb 2026

Total Runs

Tasks Tested

Total Cost

$40.18

Avg Duration

7.8m

Performance by Task

Task	Pass Rate	Avg Cost	Avg Time
dnsmasq-backdoor-detect-negative	100%	$0.35	6m
dnsmasq-backdoor-detect-negative2	100%	$0.39	7m
dropbear-brokenauth-detect-negative	100%	$0.56	9m
dropbear-brokenauth-detect-negative2	100%	$0.64	10m
ghidra-decompile-pyghidra	100%	$0.24	6m
ghidra-decompile-pyghidra-jq	100%	$0.33	10m
ghidra-decompile-vanilla	100%	$0.28	6m
lighttpd-backdoor-detect-negative	100%	$0.14	3m
lighttpd-backdoor-detect-negative2	100%	$0.17	3m
radare2-decompile	100%	$0.06	1m
radare2-decompile-jq	100%	$0.18	3m
sozu-backdoor-detect-negative	100%	$0.15	3m
sozu-backdoor-detect-negative2	100%	$0.20	4m
dropbear-brokenauth-detect-nologline	67%	$0.56	9m
lighttpd-timebomb-multiple-binaries-detect	67%	$0.45	7m
dnsmasq-backdoor-detect-obfuscated	33%	$0.54	10m
ghidra-decompile-vanilla-jq	33%	$0.43	10m
lighttpd-backdoor-multiple-binaries-detect	33%	$0.39	6m
sozu-backdoor-multiple-binaries-detect	33%	$0.68	13m
dnsmasq-backdoor-detect	0%	$0.41	7m
dnsmasq-backdoor-detect-execvp-obfuscated	0%	$0.42	9m
dnsmasq-backdoor-detect-posix-spawn	0%	$0.39	10m
dnsmasq-backdoor-detect-posix-spawn-obfuscated	0%	$0.50	9m
dnsmasq-backdoor-detect-printf	0%	$0.36	7m
dnsmasq-backdoor-detect-syscall	0%	$0.32	5m
dnsmasq-backdoor-detect-syscall-obfuscated	0%	$0.29	5m
dropbear-brokenauth-detect	0%	$0.63	14m
dropbear-brokenauth2-detect	0%	$0.69	13m
lighttpd-backdoor-detect-open	0%	$0.46	8m
lighttpd-backdoor-detect-proc-obfuscated	0%	$0.40	8m
lighttpd-backdoor-multiple-arch-binaries-detect	0%	$0.30	5m
sozu-backdoor-multiple-arch-binaries-detect	0%	$0.80	14m
sozu-timebomb-multiple-binaries-detect	0%	$0.70	12m