Vision-Language Reasoning on UHR-Micro (test)
44.1Average ScoreMAP-Agent
Evaluation results
| Method | Links | |
|---|---|---|
| MAP-Agentbase_model=Qwen3-VL-8B2026.05 | 44.1 | |
| Qwen3-VL-235B2026.05 | 32.83 | |
| Qwen3-VL-30B2026.05 | 31.51 | |
| MAP-Agentbase_model=InternVL3.5-8B2026.05 | 28.75 | |
| Qwen3-VL-8B2026.05 | 27.99 | |
| Gemini-2.5-Pro2026.05 | 27.46 | |
| DeepEyes2026.05 | 26.01 | |
| Qwen2.5-VL-7B2026.05 | 24.78 | |
| GPT-5.22026.05 | 24.76 | |
| InternVL3.5-38B2026.05 | 21.21 | |
| InternVL3.5-8B2026.05 | 20.44 | |
| Qwen2-VL-7B2026.05 | 19.35 | |
| GeoChat2026.05 | 17.17 | |
| Claude 3.7 Sonnet2026.05 | 15.37 | |
| GeoEyes2026.05 | 9.53 | |
| GeoLLaVa-8k2026.05 | 8.85 |