Your landing page’s weakest score is CTA.
How HAL (Holistic Agent Leaderboard) by Vals AI ranks
- 1.
Kimpton71
- 2.
Galileo69
- 3.
LMArena66
- 4.
HAL (Holistic Agent Leaderboard) by Vals AI61
- 5.
Scale (SEAL Evaluations)61
- 6.
Patronus AI55
CTA
44PoorCopy, size, verb, cost-feel — does the primary CTA earn the click?
Problems
- no primary CTA above the fold
- chart toggles are filters, not actions
- hero has headline + copy but no button
- no intent capture for convinced visitors
Mobile web
56FairThe phone experience as a first-class citizen.
123- News ticker banner — Ticker text is truncated mid-word ('CUA-bench: can an agent play commercial video g…') and eats top-of-fold space without a clear tap target.
- Chart filter row — PERFORMANCE SCORE / COST / LATENCY / TOKEN USAGE toggles wrap awkwardly onto two lines, reading as clutter above the chart.
- Hero body paragraph — Nine lines of dense benchmark copy push the leaderboard content deep down the phone column.
Information hierarchy
58FairDoes the design help or inhibit what the page is trying to say?
What works
What to improve
Trust
68FairProof that the product is real, competent, and safe to try.
How HAL (Holistic Agent Leaderboard) by Vals AI compares
Trust read
Value props
78FairTitle, subtitle, hero — do they sell the product's true value?
HAL (Holistic Agent Leaderboard) by Vals AI’s headline
Testing AI on Real-World Tasks
We benchmark the world's leading AI models on economically valuable tasks such as finance, software, and frontier risk like cybersecurity, recursive self improvement and mental health. We run all of our own evaluations and create many of ou
Working
Could improve