VECTOR WIREAI INTELLIGENCE
UTC
Refresh Models Deals Regulatory Sources

Claude Sonnet 5 vs Step 5 Preview

11 SHARED BENCHMARKS

Across 11 shared benchmarks, Claude Sonnet 5 scores higher on 5 and Step 5 Preview on 6. The widest gap is terminalbenchV40, where Step 5 Preview scores 33.3 against 14.1. Step 5 Preview is the cheaper of the two on tracked API pricing ($1.00 against $2.00 per million input tokens).

ANTHROPICVSSTEPFUN11 SHARED56 HEAD-TO-HEADNEWEST SCORE ADDED
AA-LCR8288.3
critpt16.920.9
gdpval5053.3
HLE57.446.5
MMMU-Pro77.376.4
scicode54.358.9
terminalbenchV4014.133.3

Best tracked score per model per benchmark (default configuration; source-attributed). ↓ marks lower-is-better metrics. Open either model for its full surface, provenance and pricing. Quoted rates are the price-setter row we currently track for each model — its direct or vendor-official listing where one exists (anthropic-official, direct), otherwise the lowest tracked offer.

Compare Claude Sonnet 5 withALL PAIRINGS →