Visual Question Answering on Human Evaluation
1,090ScoreGemini 2.5 Pro
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Gemini 2.5 ProAccess Category=API call only2026.01 | 1,090 | 1 | |
| Gemini 2.5 FlashAccess Category=API call only2026.01 | 1,082 | 2 | |
| Gemini 3 ProAccess Category=API call only2026.01 | 1,076 | 3 | |
| GPT-5 miniAccess Category=API call only2026.01 | 1,075 | 4 | |
| Molmo2-8BAccess Category=Open weights, Open data, Open code2026.01 | 1,059 | 5 | |
| Qwen3-VL-4BAccess Category=Open weights only2026.01 | 1,049 | 6 | |
| Qwen3-VL-8BAccess Category=Open weights only2026.01 | 1,048 | 7 | |
| Molmo2-4BAccess Category=Open weights, Open data, Open code2026.01 | 1,045 | 8 | |
| Molmo2-O-7BAccess Category=Open weights, Open data, Open code2026.01 | 1,034 | 9 | |
| Eagle2.5-8BAccess Category=Open weights only2026.01 | 1,022 | 10 | |
| GPT-5Access Category=API call only2026.01 | 1,019 | 11 | |
| Claude Sonnet 4.5Access Category=API call only2026.01 | 1,008 | 12 | |
| MiniCPM-V-4.5-8BAccess Category=Open weights only2026.01 | 975 | 13 | |
| VideoChat-Flash-7BAccess Category=Open models2026.01 | 959 | 14 | |
| GLM-4.1V-9BAccess Category=Open weights only2026.01 | 956 | 15 | |
| LLaVA-Video-7BAccess Category=Open models2026.01 | 955 | 16 | |
| InternVL3.5-8BAccess Category=Open weights only2026.01 | 951 | 17 | |
| Keye-VL-1.5-8BAccess Category=Open weights only2026.01 | 950 | 18 | |
| InternVL3.5-4BAccess Category=Open weights only2026.01 | 947 | 19 | |
| PLM-8BAccess Category=Open models2026.01 | 863 | 20 | |
| PLM-3BAccess Category=Open models2026.01 | 836 | 21 |