Artificial Analysis Updates Intelligence Index and Benchmarks GPT-6.1 Sol Variants

Artificial Analysis detailed the individual benchmarks of its Intelligence Index v4.3.2 and launched an evaluation page for GPT-6.1 Sol, covering five variants (max/xhigh/high/medium/low), where the model scored 52 points with per-question costs differing up to 5.5x across tiers.

2026-09-30 ~ 2026-09-30 · 2 related posts

• Artificial Analysis breaks down individual evals in Intelligence Index v4.3.2 — ArtificialAnlys · 2026-09-30
• GPT-6.1 Sol benchmarks: 5 tiers score 42-52, cost-per-task varies 5.5x — ArtificialAnlys · 2026-09-30