BinaryAudit / deepseek-v3.2

DeepSeek deepseek 39% pass rate Rank #10 of 16 MIT 1 Dec 2025

Total Runs

97

Tasks Tested

33

Total Cost

$6.16

Avg Duration

11.1m

Performance by Task

Task Pass Rate Runs Avg Cost Avg Time
dnsmasq-backdoor-detect-negative 100%
$0.03 7m
dnsmasq-backdoor-detect-negative2 100%
$0.05 10m
dropbear-brokenauth-detect-negative2 100%
$0.12 16m
GHIDRA_FAV150 ghidra-decompile-vanilla 100%
$0.03 9m
lighttpd-backdoor-detect-negative 100%
$0.02 6m
radare2-decompile 100%
$0.01 3m
radare2-decompile-jq 100%
$0.01 3m
sozu-backdoor-detect-negative 100%
$0.02 7m
sozu-backdoor-detect-negative2 100%
$0.03 7m
dnsmasq-backdoor-detect 67%
$0.05 7m
dnsmasq-backdoor-detect-obfuscated 67%
$0.06 8m
dnsmasq-backdoor-detect-syscall 67%
$0.04 6m
dropbear-brokenauth-detect-negative 67%
$0.03 8m
GHIDRA_FAV150 ghidra-decompile-pyghidra 67%
$0.02 8m
lighttpd-backdoor-detect-negative2 67%
$0.03 6m
lighttpd-backdoor-multiple-binaries-detect 33%
$0.08 13m
dnsmasq-backdoor-detect-execvp-obfuscated 0%
$0.06 12m
dnsmasq-backdoor-detect-posix-spawn 0%
$0.04 8m
dnsmasq-backdoor-detect-posix-spawn-obfuscated 0%
$0.05 9m
dnsmasq-backdoor-detect-printf 0%
$0.06 9m
dnsmasq-backdoor-detect-syscall-obfuscated 0%
$0.02 8m
dropbear-brokenauth-detect 0%
$0.16 10m
dropbear-brokenauth-detect-nologline 0%
$0.15 8m
dropbear-brokenauth2-detect 0%
$0.13 9m
GHIDRA_FAV150 ghidra-decompile-pyghidra-jq 0%
$0.05 30m
GHIDRA_FAV150 ghidra-decompile-vanilla-jq 0%
$0.08 24m
lighttpd-backdoor-detect-open 0%
$0.28 22m
lighttpd-backdoor-detect-proc-obfuscated 0%
$0.08 15m
lighttpd-backdoor-multiple-arch-binaries-detect 0%
$0.00 <1m
lighttpd-timebomb-multiple-binaries-detect 0%
$0.12 16m
sozu-backdoor-multiple-arch-binaries-detect 0%
$0.06 28m
sozu-backdoor-multiple-binaries-detect 0%
$0.06 15m
sozu-timebomb-multiple-binaries-detect 0%
$0.07 17m

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.