Clinical Triage Evaluation on MEDIQ
89.8AccuracyNumerical cutoff
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| Numerical cutoffModel=Gemma-4-31B2026.06 | 89.8 | 1,554 | 2 | |
| EvolvedModel=Gemma-4-31B2026.06 | 89.5 | 46 | 3 | |
| BinaryModel=Gemma-4-31B2026.06 | 89 | 844 | 1 | |
| ScaleModel=Gemma-4-31B2026.06 | 88.5 | 1,065 | 4 | |
| EvolvedModel=Qwen-3.5-27B2026.06 | 88.3 | 789 | 4 | |
| Numerical cutoffModel=Qwen-3.5-27B2026.06 | 88.1 | 2,214 | 2 | |
| BinaryModel=Qwen-3.5-27B2026.06 | 85.5 | 1,753 | 1 | |
| ScaleModel=Qwen-3.5-27B2026.06 | 85.1 | 1,710 | 3 | |
| ImplicitModel=Gemma-4-31B2026.06 | 80.7 | 294 | 0 | |
| EvolvedModel=Llama-3-70B2026.06 | 78.1 | 477 | 4 | |
| Numerical cutoffModel=Llama-3-70B2026.06 | 76.4 | 1,931 | 2 | |
| BinaryModel=Llama-3-70B2026.06 | 76.2 | 1,966 | 1 | |
| ScaleModel=Llama-3-70B2026.06 | 75.7 | 2,496 | 3 | |
| ImplicitModel=Llama-3-70B2026.06 | 72.4 | 418 | 0 | |
| EvolvedModel=Llama-3-8B2026.06 | 63.8 | 156 | 4 | |
| ImplicitModel=Qwen-3.5-27B2026.06 | 62.4 | 625 | 0 | |
| BinaryModel=Llama-3-8B2026.06 | 57.4 | 2,137 | 1 | |
| Numerical cutoffModel=Llama-3-8B2026.06 | 56.4 | 3,463 | 3 | |
| ScaleModel=Llama-3-8B2026.06 | 56.2 | 2,727 | 2 | |
| ImplicitModel=Llama-3-8B2026.06 | 43.6 | 154 | 0 |