Performance results of AI coding models on Altair tasks, measuring success rate and execution time with high precision.
| Model | Passed | Avg Duration | Success Rate |
|---|---|---|---|
| #1 gemini-3.5-flashNEW | 25 | 309.9s | 86% |
| #2 glm-5.2 | 11 | 320.5s | 79% |
| #3 claude-4-6-sonnet | 22 | 241.2s | 76% |
| #4 MiniMax-M3 | 8 | 244.6s | 57% |
| #5 gemini-3.1-pro | 8 | 232.8s | 53% |
| #6 glm-5.1 | 7 | 390.7s | 47% |
| #7 deepseek-4-pro | 7 | 339.9s | 47% |