deepseek-v4.1-flash
deepseek-v4.1-flash · max, showing overall score. Results for all tested efforts are available below.
Overall score67.85
deepseek-v4.1-flash · max
- Backend engineering
- 62
- Frontend interaction
- 88
- Knowledge & Reasoning
- 65
- Visual aesthetics
- 64
- Maintainability
- 55
- Evaluation time
- 35m 40s
- Reference cost
- $0.41
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 |
|---|---|---|---|---|---|---|---|---|---|
| 52.35 | 56 | 44 | 50 | 61 | 57 | 29m 47s | $0.29 | ||
| 67.85 | 62 | 88 | 65 | 64 | 55 | 35m 40s | $0.41 |
Score breakdown
- Overall score
- 67.85
- Evaluation time
- 35m 40s
- Reference cost
- $0.41
Service logic, error handling, and test verification.
Contracts and boundaries7 / 20
API contracts, input validation, and boundary behavior.
Retries and failures14 / 20
Error recovery, retry conditions, and failure handling.
CI and adversarial review13 / 20
Verification plans, risk detection, and audit evidence.
Transactions and consistency14 / 20
State transitions, concurrency control, and data consistency.
Caching and regression14 / 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 max
- Overall
- +15.5 pts 52.35 → 67.85
- Reference cost
- +$0.12 $0.29 → $0.41
- Evaluation time
- +19.7% 29m 47s → 35m 40s
Data snapshot published · 10/27/2026Scoring & cost notes