Observed arrival · 2026-10-16
EVES’s 432-question answer check
A public evaluation page for Eden’s AI verification system, comparing how often it made up answers with three models answering from memory.
○Open
⊠Login
$Paid
†Ads
✦Pretty
●Pro
◎Niche
◉Human
⚑Risk
ƒJS
- For
- People comparing AI factual-answer reliability
- Worth noticing
- EVES said it didn’t know on 20.6% of questions; the two Gemini models abstained on less than 1%.
Field notes
The test combines 300 short factual questions from SimpleQA Verified with 132 false-premise questions, including 32 written by the site. EVES returned “said it didn’t know” for 20.6% of the full set, while the two Gemini models did so on less than 1%. The page attributes about 60% of EVES’s reported gain over Claude Sonnet 5.5 to web use, and notes that no web-search-enabled comparison model was included.
Observed signals
Read the marks
Editorial observations of this landing page, not a rating.
○OpenPublic substance visible
⊠LoginAccess appeared gated
✦PrettyNotable craft visible
●ProPolished or operationally mature
◎NicheUnusually specific use
One card from the complete issue