Model comparison
Two to four models, side by side on the five measurements behind the shopping score.
Select models
2 of 4 selected| Measurement | Claude Sonnet 4.6 Anthropic · score 38 · Fair | GPT-5.5 OpenAI · score 21 · Weak |
|---|---|---|
| Checkout rate40% of score | 34.5% | 10.1% |
| Cart rate20% of score | 47.7% | 24.1% |
| Search rate10% of score | 79.8% | 70.1% |
| Turn efficiency15% of score | 44/100 | 24/100 |
| Token efficiency15% of score | 0/100 | 8/100 |
| Avg tokensper run · not in the score | 71,048 | 59,686 |
| Avg durationper run · not in the score | 33.8s | 34.4s |
Best of the selection
Checkout rateClaude Sonnet 4.634.5%
Cart rateClaude Sonnet 4.647.7%
Search rateClaude Sonnet 4.679.8%
Turn efficiencyClaude Sonnet 4.644/100
Token efficiencyGPT-5.58/100
These rates are averages across every store in our runs. Yours will differ: the same model scores differently against a strict store and an unenforced one.
Start a run