Safety on MSSBench
93.1Safety ScoreEcoAlign
Evaluation Results
| Method | Links | |
|---|---|---|
| EcoAlignCategory=Closed-source, Model=Gemini-2.5-Flash, Avg. Cost=24.12025.11 | 93.1 | |
| EcoAlignCategory=Closed-source, Model=Qwen-VL-Max, Avg. Cost=12.72025.11 | 92.8 | |
| EcoAlignCategory=Closed-source, Model=GPT-4o, Avg. Cost=21.22025.11 | 91.8 | |
| EcoAlignCategory=Open-source, Model=Llama-3.2-11B-Vision, Avg. Cost=28.22025.11 | 90.7 | |
| EcoAlignCategory=Open-source, Model=InternVL3-14B, Avg. Cost=39.32025.11 | 87.9 | |
| VLM-GuardCategory=Closed-source, Model=Qwen-VL-Max, Avg. Cost=2.22025.11 | 85 | |
| VLM-GuardCategory=Closed-source, Model=Gemini-2.5-Flash, Avg. Cost=1.42025.11 | 82.2 | |
| CoDCategory=Closed-source, Model=Qwen-VL-Max, Avg. Cost=34.32025.11 | 79.3 | |
| VLM-GuardCategory=Closed-source, Model=GPT-4o, Avg. Cost=3.12025.11 | 75.9 | |
| VLM-GuardCategory=Open-source, Model=Llama-3.2-11B-Vision, Avg. Cost=1.62025.11 | 75.8 | |
| CoDCategory=Closed-source, Model=Gemini-2.5-Flash, Avg. Cost=27.52025.11 | 75.6 | |
| VLM-GuardCategory=Open-source, Model=InternVL3-14B, Avg. Cost=1.92025.11 | 73 | |
| CoTCategory=Closed-source, Model=Gemini-2.5-Flash, Avg. Cost=86.62025.11 | 71.3 | |
| CoDCategory=Closed-source, Model=GPT-4o, Avg. Cost=25.82025.11 | 70.1 | |
| CoTCategory=Closed-source, Model=GPT-4o, Avg. Cost=104.32025.11 | 69.4 | |
| CoDCategory=Open-source, Model=InternVL3-14B, Avg. Cost=45.52025.11 | 68.4 | |
| CoTCategory=Closed-source, Model=Qwen-VL-Max, Avg. Cost=1142025.11 | 64 | |
| CoTCategory=Open-source, Model=InternVL3-14B, Avg. Cost=134.22025.11 | 62 | |
| BaseCategory=Closed-source, Model=Gemini-2.5-Flash, Avg. Cost=12025.11 | 60.5 | |
| CoDCategory=Open-source, Model=Llama-3.2-11B-Vision, Avg. Cost=362025.11 | 59.2 | |
| CoTCategory=Open-source, Model=Llama-3.2-11B-Vision, Avg. Cost=108.32025.11 | 58.9 | |
| BaseCategory=Closed-source, Model=GPT-4o, Avg. Cost=12025.11 | 58.8 | |
| BaseCategory=Closed-source, Model=Qwen-VL-Max, Avg. Cost=12025.11 | 53.4 | |
| BaseCategory=Open-source, Model=InternVL3-14B, Avg. Cost=12025.11 | 51.2 | |
| BaseCategory=Open-source, Model=Llama-3.2-11B-Vision, Avg. Cost=12025.11 | 48.4 |