Dimension Drop
Severity 10/10
2026-W12
Claude Sonnet 4.6 Stability下跌 23 分
Score Comparison
| Dimension | Previous | Current | Change |
|---|---|---|---|
| Overall (v5) | 42.0 | 53.0 | +11 |
| Code Execution (v5) | 20.8 | 59.1 | +38.3 |
| Knowledge Synthesis (v5) | 37.4 | 43.1 | +5.7 |
| Grounding (v5) | 66.7 | 76.2 | +9.5 |
| Value | 13.8 | 19.6 | +5.8 |
| Stability | 54.2 | 31.2 | -23 |
| Availability | 100.0 | 100.0 | +0 |
Affected Dimensions
Stability
Run #37 · Formula v5 · Judge v6 · Benchmark v5.1 · 2026-03-22 14:26 SGT
View Claude Sonnet 4.6 Full Profile