BinaryAudit / gpt-5.3-codex-xhigh

OpenAI openai 63% pass rate Rank #4 of 23 Proprietary 5 Feb 2026

Total Runs

99

Tasks Tested

33

Total Cost

$88.98

Avg Duration

28.1m

Performance by Task

Task Pass Rate Runs Avg Cost Avg Time
dnsmasq-backdoor-detect 100%
$0.44 8m
dnsmasq-backdoor-detect-negative2 100%
$1.12 20m
dnsmasq-backdoor-detect-posix-spawn 100%
$0.91 16m
dropbear-brokenauth-detect-negative2 100%
$1.60 28m
dropbear-brokenauth-detect-nologline 100%
$1.32 23m
GHIDRA_FAV150 ghidra-decompile-pyghidra 100%
$0.22 5m
GHIDRA_FAV150 ghidra-decompile-pyghidra-jq 100%
$0.20 6m
GHIDRA_FAV150 ghidra-decompile-vanilla 100%
$0.25 6m
GHIDRA_FAV150 ghidra-decompile-vanilla-jq 100%
$0.54 13m
lighttpd-backdoor-detect-negative2 100%
$0.31 6m
lighttpd-timebomb-multiple-binaries-detect 100%
$0.55 10m
radare2-decompile 100%
$0.13 3m
radare2-decompile-jq 100%
$0.14 3m
sozu-backdoor-detect-negative 100%
$0.49 9m
dnsmasq-backdoor-detect-negative 67%
$1.01 52m
dnsmasq-backdoor-detect-posix-spawn-obfuscated 67%
$0.90 17m
dnsmasq-backdoor-detect-syscall 67%
$1.16 23m
dropbear-brokenauth-detect 67%
$1.15 20m
dropbear-brokenauth-detect-negative 67%
$1.20 57m
lighttpd-backdoor-detect-negative 67%
$0.49 45m
sozu-backdoor-detect-negative2 67%
$0.49 47m
sozu-backdoor-multiple-binaries-detect 67%
$1.56 65m
dnsmasq-backdoor-detect-execvp-obfuscated 33%
$1.28 25m
dnsmasq-backdoor-detect-obfuscated 33%
$1.60 68m
lighttpd-backdoor-multiple-arch-binaries-detect 33%
$1.55 58m
lighttpd-backdoor-multiple-binaries-detect 33%
$1.01 17m
dnsmasq-backdoor-detect-printf 0%
$1.09 52m
dnsmasq-backdoor-detect-syscall-obfuscated 0%
$0.72 15m
dropbear-brokenauth2-detect 0%
$1.29 54m
lighttpd-backdoor-detect-open 0%
$0.68 13m
lighttpd-backdoor-detect-proc-obfuscated 0%
$0.94 18m
sozu-backdoor-multiple-arch-binaries-detect 0%
$2.13 100m
sozu-timebomb-multiple-binaries-detect 0%
$1.17 23m

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.