BinaryAudit / gpt-5.2-codex

OpenAI openai 41% pass rate Rank #9 of 16 Proprietary 18 Dec 2025

Total Runs

96

Tasks Tested

33

Total Cost

$38.07

Avg Duration

7.2m

Performance by Task

Task Pass Rate Runs Avg Cost Avg Time
dnsmasq-backdoor-detect-negative 100%
$0.27 4m
dnsmasq-backdoor-detect-negative2 100%
$0.26 4m
dropbear-brokenauth-detect-negative 100%
$0.48 7m
dropbear-brokenauth-detect-negative2 100%
$0.33 5m
GHIDRA_FAV150 ghidra-decompile-vanilla 100%
$0.57 11m
lighttpd-backdoor-detect-negative 100%
$0.15 2m
lighttpd-backdoor-detect-negative2 100%
$0.16 2m
sozu-backdoor-detect-negative 100%
$0.14 2m
sozu-backdoor-detect-negative2 100%
$0.05 1m
GHIDRA_FAV150 ghidra-decompile-pyghidra 67%
$0.30 8m
GHIDRA_FAV150 ghidra-decompile-pyghidra-jq 67%
$0.42 13m
GHIDRA_FAV150 ghidra-decompile-vanilla-jq 67%
$1.00 23m
radare2-decompile 67%
$0.03 1m
radare2-decompile-jq 67%
$0.13 4m
dnsmasq-backdoor-detect 33%
$0.36 5m
dnsmasq-backdoor-detect-execvp-obfuscated 33%
$0.28 4m
lighttpd-backdoor-multiple-binaries-detect 33%
$0.48 9m
lighttpd-timebomb-multiple-binaries-detect 33%
$0.33 7m
sozu-backdoor-multiple-binaries-detect 33%
$0.80 17m
dnsmasq-backdoor-detect-obfuscated 0%
$0.39 5m
dnsmasq-backdoor-detect-posix-spawn 0%
$0.23 3m
dnsmasq-backdoor-detect-posix-spawn-obfuscated 0%
$0.32 4m
dnsmasq-backdoor-detect-printf 0%
$0.39 5m
dnsmasq-backdoor-detect-syscall 0%
$0.29 4m
dnsmasq-backdoor-detect-syscall-obfuscated 0%
$0.30 4m
dropbear-brokenauth-detect 0%
$0.44 7m
dropbear-brokenauth-detect-nologline 0%
$0.30 4m
dropbear-brokenauth2-detect 0%
$0.33 5m
lighttpd-backdoor-detect-open 0%
$0.35 5m
lighttpd-backdoor-detect-proc-obfuscated 0%
$0.28 4m
lighttpd-backdoor-multiple-arch-binaries-detect 0%
$0.00 <1m
sozu-backdoor-multiple-arch-binaries-detect 0%
$1.06 22m
sozu-timebomb-multiple-binaries-detect 0%
$1.56 34m

All product names, logos, and brands (™/®) are the property of their respective owners; they're used here solely for identification and comparison, and their use does not imply affiliation, endorsement, or sponsorship.