Claude Opus 5 tops an Artificial Analysis coding-agent benchmark at 67

Hesamation · x · 2026-07-25

A repost reacts to an Artificial Analysis chart comparing coding-agent performance across several models and says Anthropic “really cooked” with Opus 5. The attached image shows Claude Opus 5 at the top of the chart with a composite pass@1 score of 67, tied with Codex GPT-5.6 Sol and ahead of Claude Opus 5 and other models on the same benchmark set.

Original post →

More from Models

Models channel →