AnthropicClaude Opus 4.8
Anthropic's previous flagship (Opus 4.8, superseded by Opus 5), kept as a lower-cost option. The Opus line has the lowest tool error rate in our benchmark (~2%) with efficient checkout completion (~7 turns) and a conservative, confirm-before-acting style. The model family that completed the first fully autonomous wallet-funded purchase through UCP.
- Avg tokens per run
- 84,212
- Avg duration
- 31.1s
- Turns to checkout
- 8.3
Scores come from runs on UCP stores, not estimates: every figure is a tool call, a checkout attempt or a store’s answer.
Shopping score breakdown
- Checkout rate40%22.3%
- Cart rate20%31.6%
- Search rate10%63.7%
- Turn efficiency15%46%
- Token efficiency15%0%
36.1% of runs failed with errors.
Top stores
Development stores are hidden.
| Store | Checkout | Cart |
|---|---|---|
| forever21.com | 100% | 100% |
| store hidden | 100% | 100% |
| sini.fi | 100% | 100% |
| store hidden | 70% | 90% |
| store hidden | 66.7% | 66.7% |
| store hidden | 62.5% | 62.5% |
| ucp.travel | 61.5% | 61.5% |
| houseofparfum.nl | 58.5% | 69.8% |
| store hidden | 50% | 100% |
| jbhifi.com.au | 50% | 50% |
Known issues
- LowNudge Dependency
Occasionally needs nudge (6.8%) but recovers cleanly and completes checkout.
- LowConservative Behavior
Prefers to ask for confirmation rather than proceeding autonomously. Adds turns but reduces errors.
Token usage
- Avg per run
- 84,212
- Average of all models
- 78,181
- Against the average
- +8%
- Median (p50)
- 29,197
- p90
- 164,645
Prompt and completion
Daily average tokens per run, last 30 days
Oct 29Total 37,869Prompt 37,401Completion 468
Range
- Min
- 0
- p25
- 11,938
- p50
- 29,197
- p75
- 71,598
- p90
- 164,645
- Max
- 6,246,399
Test Claude Opus 4.8 on your store
Start a run to see how Claude Opus 4.8 handles your store’s checkout, call by call.
Compare with other models
Rates are checkout rates. Full comparison view