Multidisciplinary Knowledge on MMMU
69.1ScoreGPT-4o
Evaluation Results
| Method | Links | |
|---|---|---|
| GPT-4oLLM Backbone=-, Open Data=false2025.12 | 69.1 | |
| Claude 3.5 SonnetLLM Backbone=-, Open Data=false2025.12 | 68.3 | |
| MiMo-VL-RLLLM Backbone=MiMo-7B, Open Data=false2025.12 | 66.7 | |
| Gemini-1.5 ProLLM Backbone=-, Open Data=false2025.12 | 65.8 | |
| InternVL3LLM Backbone=Qwen2.5-7B, Open Data=false2025.12 | 62.7 | |
| Qwen2.5-VLLLM Backbone=Qwen2.5-7B, Open Data=false2025.12 | 58.6 | |
| Dream-VLLLM Backbone=Dream 7B, Open Data=true2025.12 | 52.2 | |
| Dream-VLLLM Backbone=Dream 7B, Open Data=true, Input Mode=Single Image2025.12 | 51.8 | |
| DeepSeek-VL2LLM Backbone=DeepSeek-MoE 27B, Open Data=false2025.12 | 51.1 | |
| MAmmoTH-VLLLM Backbone=Qwen2.5-7B, Open Data=true2025.12 | 50.8 | |
| MAmmoTH-VLLLM Backbone=Qwen2.5-7B, Open Data=true, Input Mode=Single Image2025.12 | 49.4 | |
| Llava-CoT-11BLLM Backbone=LLaMA3.2-11B, Open Data=true2025.12 | 48.9 | |
| LLaVA-OVLLM Backbone=Qwen2-7B, Open Data=true2025.12 | 48.8 | |
| LLaDA-VLLM Backbone=LLaDA 8B, Open Data=true2025.12 | 48.6 | |
| LLaVA-OVLLM Backbone=Qwen2-7B, Open Data=true, Input Mode=Single Image2025.12 | 47.3 | |
| Molmo-8B-DLLM Backbone=Qwen2 7B, Open Data=true2025.12 | 45.3 | |
| DimpleLLM Backbone=Dream 7B, Open Data=true2025.12 | 45.2 | |
| Cambrian-1LLM Backbone=LLaMA3-8B, Open Data=true2025.12 | 42.7 | |
| LaViDa-DLLM Backbone=Dream 7B, Open Data=true2025.12 | 42.6 | |
| FUDOKILLM Backbone=Janus-1.5B, Open Data=true2025.12 | 34.7 | |
| MMaDALLM Backbone=LLaDA 8B, Open Data=false2025.12 | 30.2 |