Video Understanding on Combined Video Benchmarks
73.1Short QA Avg.GPT-5
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| GPT-5Access Type=API call only2026.01 | 73.1 | 76.3 | 70.6 | 1,031 | 10 | |
| Gemini 2.5 ProAccess Type=API call only2026.01 | 71.1 | 80.4 | 71.2 | 1,096 | 1 | |
| Gemini 3 ProAccess Type=API call only2026.01 | 71 | 78.8 | 70 | 1,082 | 3 | |
| Molmo2-8BAccess Type=Molmo2 family2026.01 | 69.9 | 64.1 | 63.1 | 1,057 | 5 | |
| Molmo2-4BAccess Type=Molmo2 family2026.01 | 69.3 | 64.5 | 62.8 | 1,041 | 8 | |
| PLM-8BAccess Type=Open models2026.01 | 68.5 | 56.2 | 56.2 | 853 | 21 | |
| Molmo2-O-7BAccess Type=Molmo2 family2026.01 | 68.1 | 59.2 | 59.7 | 1,033 | 9 | |
| Gemini 2.5 FlashAccess Type=API call only2026.01 | 67 | 74.5 | 66.7 | 1,084 | 2 | |
| Eagle2.5-8BAccess Type=Open weights only2026.01 | 67 | 65.2 | 60.7 | 1,019 | 11 | |
| GPT-5 miniAccess Type=API call only2026.01 | 66.8 | 69.8 | 65 | 1,076 | 4 | |
| VideoChat-Flash-7BAccess Type=Open models2026.01 | 66.4 | 58.1 | 56.1 | 956 | 16 | |
| PLM-3BAccess Type=Open models2026.01 | 66.3 | 53.5 | 53.9 | 841 | 20 | |
| Qwen3-VL-8BAccess Type=Open weights only2026.01 | 65.3 | 63.5 | 59.5 | 1,054 | 6 | |
| GLM-4.1V-9BAccess Type=Open weights only2026.01 | 64.2 | 60.5 | 56.9 | 962 | 14 | |
| Qwen3-VL-4BAccess Type=Open weights only2026.01 | 63.7 | 62.7 | 58.1 | 1,048 | 7 | |
| InternVL3.5-8BAccess Type=Open weights only2026.01 | 63 | 57.1 | 54.1 | 941 | 19 | |
| Claude Sonnet 4.5Access Type=API call only2026.01 | 62.8 | 66.4 | 59.6 | 1,008 | 12 | |
| MiniCPM-V-4.5-8BAccess Type=Open weights only2026.01 | 62.1 | 60.1 | 56.6 | 975 | 13 | |
| InternVL3.5-4BAccess Type=Open weights only2026.01 | 62 | 56.5 | 53.4 | 935 | 18 | |
| Keye-VL-1.5-8BAccess Type=Open weights only2026.01 | 60.1 | 60.4 | 55.7 | 952 | 17 | |
| LLaVA-Video-7BAccess Type=Open models2026.01 | 59.4 | 56.2 | 52.7 | 959 | 15 |