Kaggle RSNA Knee AUC10 - Vision (Medium): leaderboard
Vision-only knee MRI label-extraction benchmark on Kaggle (kirderf/rsna-knee-llm-label-extraction-benchmark-auc10) measuring macro ROC-AUC (%) over 12 radiology findings across 10 radiologist-labeled consensus gold studies using medium image mosaics.
Metric: Macro ROC-AUC (%). Source: www.kaggle.com. Saturation forecast: Around August 2027. 28 models tracked. Updated 21 Nov 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-6.1 Sol | 83.91 | #15 |
| 2 | GPT-5.6 Sol | 82.33 | #37 |
| 3 | GPT-6 Astra | 82.1 | #10 |
| 4 | GPT-5.5 | 79.39 | #54 |
| 5 | GPT-5.6 Terra | 71.29 | #71 |
| 6 | Claude Opus 5.5 | 70.26 | #9 |
| 7 | Gemini 4 Argon | 65.16 | #19 |
| 8 | GPT-6 Luna | 65.02 | #105 |
| 9 | Gemini 3.8 Flash | 62.77 | #43 |
| 10 | GPT-5.4 | 62.17 | #135 |
| 11 | GPT-5.6 Luna | 61.81 | #131 |
| 12 | Gemini 3.1 Flash Lite (Preview) | 60.13 | #417 |
| 13 | GPT-OSS-20B | 57.63 | #1183 |
| 14 | Gemma 4 31B | 56.83 | #382 |
| 15 | Claude Sonnet 5.5 | 56.83 | #24 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=kaggle-rsna-knee-auc10-vision-medium · How It Works · Data refreshed daily, snapshot 2026-11-21.