Quality Scoring on Software Scenario
100Average ScoreForce Strong
Evaluation results
| Method | Links | |
|---|---|---|
| Force StrongModel=claude2026.01 | 100 | |
| Force StrongModel=gemini2026.01 | 100 | |
| CASTERModel=gemini2026.01 | 100 | |
| CASTERModel=qwen2026.01 | 100 | |
| Force WeakModel=gemini2026.01 | 99.8 | |
| Force StrongModel=qwen2026.01 | 99.3 | |
| Force StrongModel=deepseek2026.01 | 98.6 | |
| Force WeakModel=deepseek2026.01 | 98.3 | |
| CASTERModel=deepseek2026.01 | 98.1 | |
| Force WeakModel=qwen2026.01 | 97.8 | |
| CASTERModel=openai2026.01 | 97 | |
| CASTERModel=claude2026.01 | 96.4 | |
| Force StrongModel=openai2026.01 | 95.3 | |
| Force WeakModel=claude2026.01 | 93.3 | |
| Force WeakModel=openai2026.01 | 82.2 |