Visual Question Answering on MuirBench
78.6AccuracyGPT-5
Evaluation Results
| Method | Links | |
|---|---|---|
| GPT-5Access=API call only2026.03 | 78.6 | |
| GLM-4.1V-9BAccess=Open weights only2026.03 | 74.7 | |
| Gemini 2.5 ProAccess=API call only2026.03 | 74.5 | |
| Gemini 2.5 FlashAccess=API call only2026.03 | 73.5 | |
| GPT-5 miniAccess=API call only2026.03 | 71.4 | |
| Qwen3-VL-8BAccess=Open weights only2026.03 | 64.4 | |
| Qwen3-VL-4BAccess=Open weights only2026.03 | 63.8 | |
| Molmo2-8BAccess=Fully open2026.03 | 63.7 | |
| MolmoPoint-8BAccess=MolmoPoint2026.03 | 62.5 | |
| Eagle2.5-8BAccess=Open weights only2026.03 | 61.8 | |
| Molmo2-4BAccess=Fully open2026.03 | 60.5 | |
| Claude Sonnet 4.5Access=API call only2026.03 | 59.6 | |
| MolmoPoint-8B-O-7BAccess=Fully open2026.03 | 58.4 | |
| InternVL3.5-8BAccess=Open weights only2026.03 | 55.8 | |
| MiniCPM-V-4.5-8BAccess=Open weights only2026.03 | 53.3 | |
| InternVL3.5-4BAccess=Open weights only2026.03 | 53.1 | |
| Keye-VL-1.5-8BAccess=Open weights only2026.03 | 51.2 | |
| PLM-3BAccess=Fully open2026.03 | 25.7 | |
| PLM-8BAccess=Fully open2026.03 | 23.5 |