Multimodal Logical Reasoning on LogicVista (test)
64.4ScoreGPT-4o
Evaluation Results
| Method | Links | |
|---|---|---|
| GPT-4oSize=-2026.03 | 64.4 | |
| Open-Vision-ReasonerSize=7B2026.03 | 54.8 | |
| Insight-V++ (Qwen2.5-VL + Self-Evolving)Size=7B2026.03 | 52.9 | |
| Insight-V (Qwen2.5-VL + Multi-Agent RL)Size=7B2026.03 | 51.7 | |
| OpenMMReasonerSize=7B2026.03 | 50 | |
| MMR1Size=7B2026.03 | 48.9 | |
| VLAA-Thinker-Qwen2.5Size=7B2026.03 | 48.5 | |
| Qwen2.5-VLSize=7B2026.03 | 47.9 | |
| MM-Eureka-QwenSize=7B2026.03 | 46.3 | |
| LLaVA-OneVision-1.5Size=8B2026.03 | 45.8 | |
| OpenVLThinkerSize=7B2026.03 | 44.5 | |
| InternVL3-8BSize=8B2026.03 | 43.6 | |
| ThinkLite-VLSize=7B2026.03 | 42.7 | |
| VL-RethinkerSize=7B2026.03 | 42.7 | |
| GPT-4o miniSize=-2026.03 | 41.4 | |
| LLaVA-OneVisionSize=7B2026.03 | 32 |