The August 2026 AI Intelligence Index rankings are out, and Anthropic's Claude Opus 5 holds the top spot with 63 points. Right behind it, the race is extremely close.
What Happened
The new leaderboard puts five models within three points of each other:
Claude Opus 5 (max) — 63 pts
Claude Fable 5 (with fallback) — 62 pts
GPT-5.6 Sol (max) — 61 pts
Grok 4.6 (high) — 61 pts
Kimi K3 (max) — 60 pts
Notable mentions: Qwen3.8 Max (58), Muse Spark 1.2 and GPT-5.6 Terra (57), Grok 4.5 (56), and Claude Sonnet 5 (55).
Why It Matters
A three-point gap at the top means raw intelligence scores are not separating these models by much. When you choose a daily driver for coding or complex tasks, the benchmark alone probably will not settle it — tooling, speed, and price will matter more than the index.
Comments
Join the conversation — sign in to comment.
No comments yet — start the conversation!