Skip to main content
How it works
Benchmarks
Blog
About us
Token economics
Talk to the founder
Menu
How it works
Benchmarks
Blog
About us
Token economics
Talk to the founder
BinaryAudit
/
Claude Sonnet 4
Anthropic
anthropic
36% pass rate
Rank #22 of 26
Proprietary
22 May 2025
Total Runs
97
Tasks Tested
33
Total Cost
$164.38
Avg Duration
5.5m
Performance by Task
Task
Pass Rate
Runs
Avg Cost
Avg Time
dnsmasq-backdoor-detect
100%
$0.32
3m
dnsmasq-backdoor-detect-negative
100%
$0.36
4m
dnsmasq-backdoor-detect-negative2
100%
$0.45
4m
dropbear-brokenauth-detect-negative
100%
$0.52
5m
lighttpd-backdoor-detect-negative2
100%
$0.23
3m
radare2-decompile
100%
$0.09
1m
radare2-decompile-jq
100%
$0.23
2m
sozu-backdoor-detect-negative
100%
$0.22
3m
GHIDRA_FAV150
ghidra-decompile-pyghidra
67%
$0.32
6m
lighttpd-backdoor-detect-negative
67%
$0.25
2m
dnsmasq-backdoor-detect-posix-spawn
33%
$0.50
5m
dnsmasq-backdoor-detect-printf
33%
$0.25
3m
dnsmasq-backdoor-detect-syscall
33%
$0.47
5m
dnsmasq-backdoor-detect-syscall-obfuscated
33%
$0.37
4m
dropbear-brokenauth-detect-negative2
33%
$0.50
4m
GHIDRA_FAV150
ghidra-decompile-vanilla
33%
$0.36
6m
sozu-backdoor-multiple-arch-binaries-detect
33%
$0.25
3m
dnsmasq-backdoor-detect-execvp-obfuscated
0%
$0.37
4m
dnsmasq-backdoor-detect-obfuscated
0%
$0.45
4m
dnsmasq-backdoor-detect-posix-spawn-obfuscated
0%
$0.39
4m
dropbear-brokenauth-detect
0%
$0.59
4m
dropbear-brokenauth-detect-nologline
0%
$0.44
4m
dropbear-brokenauth2-detect
0%
$0.54
4m
GHIDRA_FAV150
ghidra-decompile-pyghidra-jq
0%
$0.62
12m
GHIDRA_FAV150
ghidra-decompile-vanilla-jq
0%
$0.54
8m
lighttpd-backdoor-detect-open
0%
$0.39
4m
lighttpd-backdoor-detect-proc-obfuscated
0%
$0.47
4m
lighttpd-backdoor-multiple-arch-binaries-detect
0%
$0.00
<1m
lighttpd-backdoor-multiple-binaries-detect
0%
$0.64
5m
lighttpd-timebomb-multiple-binaries-detect
0%
$0.90
7m
sozu-backdoor-detect-negative2
0%
$0.40
4m
sozu-backdoor-multiple-binaries-detect
0%
$0.42
3m
sozu-timebomb-multiple-binaries-detect
0%
$42.21
45m