gpt-6POST
GPT-6 Sol tops Vercel's security bug benchmark
GPT-6 Sol reportedly scored higher than Astra on Vercel's security bug benchmark at the xhigh setting while costing one fifth as much to run, so the poster suggests trying it first for code audits.
🥔🥔🥔
@argofowlArchived 2026.09.27Originally posted 2026.09.27
🥔🥔🥔 @argofowlOriginally posted 2026.09.27
gpt-6 sol just topped vercel's security bug benchmark 👀 at xhigh, it beat astra's score for one fifth of the run cost i'd try sol first for code audits https://x.com/vercel_dev/status/2103594322683912635
This post quotes another post: “GPT-6 Sol tops DeepSecBench Pareto frontier with 80% lower run cost”
x.com/argofowl/status/2103922300852863470Archived 2026.09.27 09:53