Observed arrival · 2026-11-19
FrontierEE Bench puts AI models at the electrical lab bench
A public benchmark that tests frontier AI models on hands-on electrical-engineering tasks using real lab instruments.
○Open
⊠Login
$Paid
†Ads
✦Pretty
●Pro
◎Niche
◉Human
⚑Risk
ƒJS
- For
- Researchers evaluating AI on physical engineering work
- Worth noticing
- The page says its 95% credible intervals overlap, so the displayed cost-performance frontier may shift with more trials.
Field notes
Each model was run three times on each of the 10 tasks, and the page reports mean API spend across those runs. Task descriptions include estimating a diode’s junction temperature, accounting for foam that affects a sump sensor, and separating a motor’s winding properties from brush-contact and heating effects. The page also provides four example task videos.
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
ƒJavaScriptBrowser-side code central
One card from the complete issue