Information Analysis and Processing on MARS Level 3 (Evaluation Set)
1.9Avg Rank (AI GPT-5)Ours
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| OursParam.=1800B2025.11 | 1.9 | 1.3 | |
| Qwen3-235B-AParam.=22B2025.11 | 2.3 | 4.4 | |
| Llama-4-ScoutParam.=17B2025.11 | 4.5 | 6.4 | |
| Llama-VisionParam.=11B2025.11 | 5.5 | 6.2 | |
| Qwen-VLParam.=7B2025.11 | 6.2 | 7 | |
| Llama-3.2-11BParam.=11B2025.11 | 6.4 | 5.8 | |
| LLaVA-1.5Param.=7B2025.11 | 6.8 | 5.6 | |
| GPT-ossParam.=20B2025.11 | 7.3 | 6.6 | |
| Llama-3.2-90BParam.=90B2025.11 | 7.4 | 6.8 | |
| Qwen2-VLParam.=72B2025.11 | 8.5 | 7.7 | |
| Qwen2.5-VLParam.=72B2025.11 | 9.2 | 8.2 |