HuatuoGPT-Vision-7B: benchmark results
Provider: FreedomIntelligence. Released 2024-06-28. Access: Open.
Aggregate Rating 1521 ± 19, rank #2054 of 10916 rated models, from 120 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| BrainTRACE - Differential | 14.1 | Clinical rubric pass (%) | 100 |
| BrainTRACE - L1 Abnormality | 23.5 | Clinical rubric pass (%) | 88.9 |
| BrainTRACE - L4 Trajectory | 7.6 | Clinical rubric pass (%) | 88.9 |
| PathBind - Text-only PathMMU Test-tiny | 45.7 | Score (%) | 79.4 |
| BrainTRACE - L2 Appearance | 15.5 | Clinical rubric pass (%) | 77.8 |
| SurgCoT - Affordance Mapping | 67.24 | Affordance mapping main-question accuracy (%) on SurgCoT sur | 77.8 |
| SurgCoT - Anomaly Onset Tracking | 63.89 | Anomaly onset tracking main-question accuracy (%) on SurgCoT | 77.8 |
| PathBind - Text-only PathMMU Test-all | 44.1 | Score (%) | 76.5 |
| PathBind - Text-only Legacy Average | 40.4 | Score (%) | 70.6 |
| PathBind - Text-only VQA D6 | 30.4 | Score (%) | 70.6 |
| PathBind - Image VQA D3 | 48.8 | Score (%) | 67.6 |
| Holtercare-Bench - Heart Rate Extremum Timing (Video) | 36.31 | Accuracy (%; multiple-choice closed QA on the millisecond ti | 66.7 |
Interactive version: theaggregate.ai/model?slug=huatuogpt-vision-7b · How It Works · Data refreshed daily, snapshot 2026-11-08.