General Multimodal Intelligence Evaluation on Internal Benchmark (test)
61.6Overall ScoreGemini 2.5 Pro
Evaluation Results
| Method | Links | ||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Gemini 2.5 Prothinking=true2025.05 | 61.6 | 64.4 | 74.8 | 70.7 | 54.9 | 43.1 | 64 | 73.6 | 52.4 | 75.5 | 63.1 | 69.2 | 43.7 | 54.7 | |
| Seed1.5-VLthinking=true2025.05 | 59.3 | 63 | 72.4 | 67.2 | 47.7 | 44.1 | 63.3 | 64.9 | 47.6 | 73.1 | 63.1 | 69.6 | 44 | 47.1 | |
| OpenAI o4-minitool use=false2025.05 | 55.4 | 56.8 | 64.8 | 64.4 | 27.6 | 38.4 | 55 | 57.8 | 57.4 | 70.9 | 52.9 | 68.7 | 60.6 | 39.8 | |
| OpenAI o1thinking=true2025.05 | 54 | 51.6 | 74.5 | 55.7 | 43.6 | 42.3 | 56.1 | 68.5 | 44.9 | 66.3 | 53.2 | 63.8 | 39.9 | 30.2 | |
| Claude 3.7 Sonnetthinking=true2025.05 | 48.6 | 48.4 | 55.7 | 57.1 | 34.1 | 35.9 | 45.2 | 50.8 | 39.6 | 64.7 | 53.2 | 50.5 | 54.6 | 29.1 |