BinaryAudit / gemini-3-flash-preview

google 48% pass rate Rank #5 of 16 Proprietary 17 Dec 2025

Total Runs

Tasks Tested

Total Cost

$25.00

Avg Duration

5.7m

Performance by Task

Task	Pass Rate	Avg Cost	Avg Time
dnsmasq-backdoor-detect-negative2	100%	$0.43	6m
dnsmasq-backdoor-detect-obfuscated	100%	$0.18	5m
dnsmasq-backdoor-detect-posix-spawn	100%	$0.16	3m
dnsmasq-backdoor-detect-syscall	100%	$0.09	2m
ghidra-decompile-vanilla	100%	$0.06	3m
lighttpd-backdoor-detect-negative2	100%	$0.13	2m
sozu-backdoor-detect-negative	100%	$0.11	4m
sozu-backdoor-detect-negative2	100%	$0.13	3m
dnsmasq-backdoor-detect	67%	$0.18	3m
dropbear-brokenauth-detect	67%	$0.40	5m
dropbear-brokenauth-detect-negative2	67%	$0.41	5m
dropbear-brokenauth-detect-nologline	67%	$0.34	4m
ghidra-decompile-vanilla-jq	67%	$0.16	22m
lighttpd-backdoor-detect-negative	67%	$0.11	2m
radare2-decompile	67%	$0.01	1m
radare2-decompile-jq	67%	$0.03	1m
dnsmasq-backdoor-detect-execvp-obfuscated	33%	$0.35	5m
dnsmasq-backdoor-detect-negative	33%	$0.53	8m
dnsmasq-backdoor-detect-printf	33%	$0.40	6m
dropbear-brokenauth-detect-negative	33%	$0.46	10m
ghidra-decompile-pyghidra	33%	$0.02	2m
ghidra-decompile-pyghidra-jq	33%	$0.05	8m
lighttpd-backdoor-detect-proc-obfuscated	33%	$0.40	5m
lighttpd-backdoor-multiple-binaries-detect	33%	$0.33	8m
lighttpd-timebomb-multiple-binaries-detect	33%	$0.11	4m
sozu-backdoor-multiple-arch-binaries-detect	33%	$0.17	6m
dnsmasq-backdoor-detect-posix-spawn-obfuscated	0%	$0.54	7m
dnsmasq-backdoor-detect-syscall-obfuscated	0%	$0.58	11m
dropbear-brokenauth2-detect	0%	$0.70	6m
lighttpd-backdoor-detect-open	0%	$0.64	7m
lighttpd-backdoor-multiple-arch-binaries-detect	0%	$0.00	<1m
sozu-backdoor-multiple-binaries-detect	0%	$0.12	6m
sozu-timebomb-multiple-binaries-detect	0%	$0.45	16m