Performance results of AI coding models on gel tasks, measuring success rate and execution time with high precision.
| Model | Passed | Avg Duration | Success Rate |
|---|---|---|---|
| #1 claude-5-sonnetNEW | 11 | 570.6s | 85% |
| #2 gemini-3.5-flash | 8 | 686.0s | 62% |
| #3 deepseek-4-pro | 4 | 996.0s | 31% |
| #4 glm-5.2 | 2 | 1048.2s | 15% |