Benchmark chart shows Claude Opus 5 ahead on coding, search, and biology tasks

legit_api · x · 2026-07-25

A repost highlights a benchmark chart for Claude Opus 5, showing the model stacked against Fable 5, Opus 4.8, and GPT-5.6 Sol across agentic coding, knowledge work, search, computer use, workflows, health, and biology.

The image suggests Opus 5 is a significant upgrade, especially on agentic coding and several reasoning-style tasks, though some task rows still show close competition with other frontier models.

Related event: Anthropic Unveils Claude Opus 5 with Top Benchmark Scores(28 posts)→

Original post →

More from Models

Models channel →