Running a lower-cost model multiple times can match the vulnerability detection rate of…
This is a dev post classified by AGK Intelligence as Security (a free resource), kept by the Dev Radar because it carries real work, not commentary.
Running a lower-cost model multiple times can match the vulnerability detection rate of frontier models at a small fraction of the cost. On a 30-task subset of dfbench, five runs of DeepSeek v4.1 Flash found about 71% of known vulnerabilities, up from about 52% with a single run, at $4.24 per task. Running it 15 times increased recall further to 80%, roughly 13% of Mythos 5’s cost, which found approximately 69% of the vulnerabilities. Because LLMs are stochastic, they can find different vulnerabilities each time they analyze the same codebase. Combining findings across runs increases coverag
Posted by depthfirst (2k followers) 1 h ago · 23 likes · 802 views · view the original post on X. Kept by the Dev Radar as Security.
More dev work like this
- HIRING: FASTEST GROWING CYBERSECURITY STARTUPS (EARLY STAGE) — @clairejyz
- I'm gonna start Tip of the week series, an educational series where i can give you a… — @m4rio_eth
- I got access to the model hacking everything and in just 30 mins it hacked me — @Rasmic
- Open-sourcing the AI harness built for security research. — @0x0SojalSec
- I built agent-keychain — @stevederico
- What happens when an AI agent encounters malicious instructions? 🛡️ — @windowsdev
- i'm a coffee shop larper, which means my claude sits open on a public table every time i… — @MKBuildsDev
- Mistral’s agents ALSO escaped their sandboxes. — @Hesamation
Every post is read and classified by AGK Intelligence: what it is, which market it belongs to, and whether the link is a real tool. 31.5k posts from 5k X accounts over the last 21 days, 9.1k tools, 12 markets. New posts are collected several times an hour and the full ranking is recomputed every hour — last update 2026-11-03 02:07 UTC. Full method.