Visual Question Answering on AI2D 65 (test)
98.7ScoreGemini 3 Pro
Evaluation results
| Method | Links | |
|---|---|---|
| Gemini 3 ProModel Category=API call only2026.01 | 98.7 | |
| GPT-5Model Category=API call only2026.01 | 97.1 | |
| Molmo-72BModel Category=Molmo1 family2026.01 | 96.3 | |
| Gemini 2.5 FlashModel Category=API call only2026.01 | 95.9 | |
| GPT-5 miniModel Category=API call only2026.01 | 95.8 | |
| Molmo2-8BModel Category=Molmo2 family2026.01 | 95.8 | |
| Molmo2-4BModel Category=Molmo2 family2026.01 | 95.6 | |
| Gemini 2.5 ProModel Category=API call only2026.01 | 94.3 | |
| Molmo2-O-7BModel Category=Molmo2 family2026.01 | 93.7 | |
| Molmo-7B-DModel Category=Molmo1 family2026.01 | 93.2 | |
| PLM-8BModel Category=Open models2026.01 | 92.7 | |
| Claude Sonnet 4.5Model Category=API call only2026.01 | 91.5 | |
| PLM-3BModel Category=Open models2026.01 | 90.9 | |
| Molmo-7B-OModel Category=Molmo1 family2026.01 | 90.7 | |
| Keye-VL-1.5-8BModel Category=Open weights only2026.01 | 89.5 | |
| GLM-4.1V-9BModel Category=Open weights only2026.01 | 87.9 | |
| MiniCPM-V-4.5-8BModel Category=Open weights only2026.01 | 86.5 | |
| MolmoE-1BModel Category=Molmo1 family2026.01 | 86.4 | |
| Qwen3-VL-8BModel Category=Open weights only2026.01 | 85.7 | |
| Eagle2.5-8BModel Category=Open weights only2026.01 | 84.5 | |
| Qwen3-VL-4BModel Category=Open weights only2026.01 | 84.1 | |
| InternVL3.5-8BModel Category=Open weights only2026.01 | 84 | |
| InternVL3.5-4BModel Category=Open weights only2026.01 | 82.6 |