claude-opus-4-8
claude-opus-4-8 · xhigh, showing overall score. Results for all tested efforts are available below.
Overall score76.80
claude-opus-4-8 · xhigh
- Backend engineering
- 61
- Frontend interaction
- 87
- Knowledge & Reasoning
- 90
- Visual aesthetics
- 68
- Maintainability
- 55
- Evaluation time
- 102m 10s
- Reference cost
- $9.18
Complete evaluation · API reference cost
Results by reasoning effort
Swipe to see all results →
| Model / effort | Overall | Backend engineering | Frontend interaction | Knowledge & Reasoning | Visual aesthetics | Maintainability | Total eval. time | Total ref. cost | Compare |
|---|---|---|---|---|---|---|---|---|---|
| 65.65 | 48 | 72 | 75 | 66 | 55 | 84m 26s | $4.40 | ||
| 76.80 | 61 | 87 | 90 | 68 | 55 | 102m 10s | $9.18 | ||
| 75.85 | 64 | 88 | 85 | 68 | 55 | 106m 52s | $16.33 |
Score breakdown
- Overall score
- 76.80
- Evaluation time
- 102m 10s
- Reference cost
- $9.18
Service logic, error handling, and test verification.
Contracts and boundaries7 / 20
API contracts, input validation, and boundary behavior.
Retries and failures16 / 20
Error recovery, retry conditions, and failure handling.
CI and adversarial review11 / 20
Verification plans, risk detection, and audit evidence.
Transactions and consistency15 / 20
State transitions, concurrency control, and data consistency.
Caching and regression12 / 20
Cache invalidation, data updates, and regression coverage.
What changes with higher reasoning effort
Compare overall scores, cost, and time across neighboring tested efforts.
high xhigh
- Overall
- +11.2 pts 65.65 → 76.80
- Reference cost
- +$4.77 $4.40 → $9.18
- Evaluation time
- +21.0% 84m 26s → 102m 10s
Data snapshot published · 10/27/2026Scoring & cost notes