Model results · Anthropic

Claude Sonnet 5

Every published VulcanBench measurement of Claude Sonnet 5, newest first: accuracy, cost, tokens, and time, graded by deterministic hidden tests on real tasks. For the newer Claude Sonnet 5.5, see its own page.

Report03

Claude Fable 5 vs Sonnet 5 vs GLM-5.2 benchmark

Accuracy converges across three models while efficiency spans an order of magnitude. Fable 5 at low effort matches Sonnet 5 at high effort for a quarter of the cost.

2026-07-01

→

Report01

Claude Sonnet 5 vs Opus 4.8 benchmark

The pilot run that established cost as the live axis: both frontier models solve every real, decontaminated bug, and the gap is entirely economic.

2026-07-01

→

Report02

Claude Sonnet 5 vs Opus 4.8 reasoning-effort benchmark

Does more reasoning effort pay off? A 936-run sweep of Sonnet 5 and Opus 4.8 across low, medium, and high effort.

2026-06-30

→

How would Claude Sonnet 5 do on your stack?

Public results answer "which model is stronger on this suite." Whether Claude Sonnet 5 belongs in your routing table depends on your languages, your codebases, and your task mix, that gets measured, not guessed. Grading rules are in the methodology.

Optimize your model routing All reports