| Model,Model Type,Vendor,Overall Accuracy,Precision,Recall,F1 Score,Over-redaction Rate,Processing Time (s),Cost per Document ($),Healthcare Accuracy,Financial Accuracy,Government Accuracy,Legal Accuracy,Personal Accuracy |
| GPT-4o,Proprietary,OpenAI,0.941,0.945,0.938,0.941,0.023,2.3,0.012,0.952,0.938,0.933,0.941,0.940 |
| Claude-3.5-Sonnet,Proprietary,Anthropic,0.928,0.932,0.924,0.928,0.031,3.1,0.015,0.939,0.925,0.920,0.928,0.927 |
| Gemini-1.5-Pro,Proprietary,Google,0.915,0.919,0.911,0.915,0.038,2.8,0.008,0.926,0.912,0.907,0.915,0.914 |
| LLaMA-3.1-70B,Open Source,Meta,0.882,0.887,0.877,0.882,0.052,4.2,0.003,0.893,0.879,0.874,0.882,0.881 |
| Mistral-Large,Proprietary,Mistral AI,0.871,0.875,0.867,0.871,0.048,3.7,0.011,0.882,0.868,0.863,0.871,0.870 |
| GPT-4o-mini,Proprietary,OpenAI,0.856,0.860,0.852,0.856,0.061,1.8,0.002,0.867,0.853,0.848,0.856,0.855 |
| Claude-3-Haiku,Proprietary,Anthropic,0.834,0.838,0.830,0.834,0.078,2.1,0.006,0.845,0.831,0.826,0.834,0.833 |
| Gemini-1.5-Flash,Proprietary,Google,0.821,0.825,0.817,0.821,0.085,2.4,0.004,0.832,0.818,0.813,0.821,0.820 |
