EyeBench-v3 Updates: Sol Model Takes the Lead

EyeBench-v3 added three new models: Sol, Terra, and Luna. Sol topped the leaderboard with a 41% score in a challenging geographic image localization benchmark, while Terra and Luna showed relatively modest performance.

2026-07-10 ~ 2026-07-10 · 4 related posts

• Three New Models Added to eyebench-v3 — adonis_singh · 2026-07-10
• EyeBench-v3 Leaderboard Update: Sol Takes the Crown — adonis_singh · 2026-07-10
• GPT-5.6 Additional Visual Benchmark Results — Afinetheorem · 2026-07-10
• Evaluating Multimodal Models on Geospatial Image Localization — Afinetheorem · 2026-07-10