BinaryAudit / gpt-5.4-high

OpenAI openai 49% pass rate Rank #8 of 26 Proprietary 5 Mar 2026

Total Runs

99

Tasks Tested

33

Total Cost

$80.87

Avg Duration

13.9m

Performance by Task

Task Pass Rate Runs Avg Cost Avg Time
dnsmasq-backdoor-detect-negative 100%
$0.62 10m
dnsmasq-backdoor-detect-negative2 100%
$0.58 9m
dropbear-brokenauth-detect-negative 100%
$0.70 8m
dropbear-brokenauth-detect-negative2 100%
$0.77 9m
GHIDRA_FAV150 ghidra-decompile-pyghidra 100%
$0.20 4m
GHIDRA_FAV150 ghidra-decompile-pyghidra-jq 100%
$0.30 8m
GHIDRA_FAV150 ghidra-decompile-vanilla 100%
$0.47 7m
lighttpd-backdoor-detect-negative 100%
$0.39 5m
lighttpd-backdoor-detect-negative2 100%
$0.25 3m
radare2-decompile 100%
$0.09 1m
radare2-decompile-jq 100%
$0.17 3m
sozu-backdoor-detect-negative 100%
$0.33 4m
sozu-backdoor-detect-negative2 100%
$0.25 3m
dnsmasq-backdoor-detect-obfuscated 67%
$0.83 11m
dnsmasq-backdoor-detect-posix-spawn 67%
$0.74 10m
GHIDRA_FAV150 ghidra-decompile-vanilla-jq 67%
$0.54 9m
dnsmasq-backdoor-detect-printf 33%
$0.56 8m
dnsmasq-backdoor-detect-syscall 33%
$1.36 18m
lighttpd-backdoor-multiple-binaries-detect 33%
$0.91 12m
lighttpd-timebomb-multiple-binaries-detect 33%
$2.32 48m
dnsmasq-backdoor-detect 0%
$0.64 10m
dnsmasq-backdoor-detect-execvp-obfuscated 0%
$0.96 13m
dnsmasq-backdoor-detect-posix-spawn-obfuscated 0%
$0.45 6m
dnsmasq-backdoor-detect-syscall-obfuscated 0%
$0.68 9m
dropbear-brokenauth-detect 0%
$0.83 11m
dropbear-brokenauth-detect-nologline 0%
$0.72 9m
dropbear-brokenauth2-detect 0%
$0.72 8m
lighttpd-backdoor-detect-open 0%
$0.44 6m
lighttpd-backdoor-detect-proc-obfuscated 0%
$0.56 7m
lighttpd-backdoor-multiple-arch-binaries-detect 0%
$0.51 7m
sozu-backdoor-multiple-arch-binaries-detect 0%
$2.14 58m
sozu-backdoor-multiple-binaries-detect 0%
$2.10 52m
sozu-timebomb-multiple-binaries-detect 0%
$3.84 69m

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.