Observed arrival · 2026-09-25

SUPINT: a firmware bug hunt for frontier models

supint.org Visit website

Editorial interest 82/100 Selection signal · not a rating of the site

SUPINT publishes benchmarks of AI models tackling engineering tasks, with grades and claim checks tied to the code under review.

Landing page captured for the 2026-09-25 issue.

For: AI evaluators and radio-firmware engineers
Worth noticing: The benchmark uses a DM-1701/HR-C6000 radio-firmware task and reports claims held, wrong claims, and decode-critical issues.

Field notes

The brief names specific hardware and software context: DM-1701, HR-C6000, AT1846S, STM32F405, and P25. The leaderboard separates weighted score from six rubric dimensions and also shows claims held, wrong claims, and decode-critical issues; the page labels the first benchmark retired.

Observed signals

Read the marks

Editorial observations of this landing page, not a rating.

○OpenPublic substance visible

✦PrettyNotable craft visible

●ProPolished or operationally mature

◎NicheUnusually specific use

One card from the complete issue

Vandalism in Demo Mode

315,813 arrived 1,000 judged 972 catalogued Enter the complete issue