glm-5.3-flash
glm-5.3-flash · high, showing overall score. Results for all tested efforts are available below.
Overall score49.60
glm-5.3-flash · high
- Backend engineering
- 55
- Frontend interaction
- 76
- Knowledge & Reasoning
- 20
- Visual aesthetics
- 70
- Maintainability
- 59
- Evaluation time
- 91m 45s
- Reference cost
- $0.079Partial cost
Complete evaluation · API reference cost
Backend tested · 10/9/2026Tasks & methodology
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 |
|---|---|---|---|---|---|---|---|---|---|
| 49.60 | 55 | 76 | 20 | 70 | 59 | 91m 45s | $0.079Partial cost | ||
| 56.80 | 54 | 75 | 40 | 78 | 53 | 102m 55s | $0.081Partial cost |
Score breakdown
- Overall score
- 49.60
- Evaluation time
- 91m 45s
- Reference cost
- $0.079
Service logic, error handling, and test verification.
Contracts and boundaries6 / 20
API contracts, input validation, and boundary behavior.
Retries and failures12 / 20
Error recovery, retry conditions, and failure handling.
CI and adversarial review13 / 20
Verification plans, risk detection, and audit evidence.
Transactions and consistency10 / 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
- +7.2 pts 49.60 → 56.80
- Reference cost
- — — → —
- Evaluation time
- +12.2% 91m 45s → 102m 55s
Data snapshot published · 10/27/2026Scoring & cost notes