HealthGPT-Pro-8B: benchmark results
Provider: Other.
Aggregate Rating 1528 ± 50, rank #1971 of 10916 rated models, from 12 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| ObGynLongBench - History-Level EHR - L1 Single Evidence | 67.2 | Accuracy (%) | 100 |
| ObGynLongBench - History-Level EHR | 61.3 | Accuracy (%; full pre-decision EHR history) | 75 |
| ObGynLongBench - History-Level EHR - L3 Long-Horizon Reasoning | 57.3 | Accuracy (%) | 59.4 |
| ObGynLongBench - Visit-Level EHR | 58.1 | Accuracy (%; same-day visit records) | 56.2 |
| ObGynLongBench - History-Level EHR - L2 Multi-Source Integration | 59.8 | Accuracy (%) | 50 |
| ObGynLongBench - Evidence-Only | 73.5 | Accuracy (%; supporting evidence only) | 43.8 |
| MedPIC-Bench - Guideline-Following | 51.8 | Accuracy (%) | 29.6 |
| MedPIC-Bench - Overall | 42.6 | Accuracy (%) | 22.2 |
| MedPIC-Bench - Rule Deactivation | 20.6 | Accuracy (%) | 16.7 |
| MedPIC-Bench - Rule Activation | 42.3 | Accuracy (%) | 11.1 |
| MedPIC-Bench - Counterfactual Pair | 5.6 | Both-correct pair rate (%) | 9.3 |
| MedPIC-Bench - Counterfactual | 28.4 | Accuracy (%) | 0 |
Interactive version: theaggregate.ai/model?slug=healthgpt-pro-8b · How It Works · Data refreshed daily, snapshot 2026-11-08.