Kaggle RSNA Knee AUC10 - Vision (Small): 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 small image mosaics.
Metric: Macro ROC-AUC (%). Source: www.kaggle.com. Saturation forecast: Around June 2027. 26 models tracked. Updated 21 Nov 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-6.1 Sol | 86.7 | #15 |
| 2 | GPT-5.6 Terra | 78.17 | #71 |
| 3 | GPT-5.6 Sol | 76.48 | #37 |
| 4 | GPT-5.6 Luna | 73.19 | #131 |
| 5 | GPT-5.4 | 68.82 | #135 |
| 6 | GPT-5.5 | 68.14 | #54 |
| 7 | Gemini 4 Argon | 65.93 | #19 |
| 8 | Claude Opus 5.5 | 65.71 | #9 |
| 9 | Gemini 3.1 Flash Lite (Preview) | 65.56 | #417 |
| 10 | Gemini 3.1 Pro (Preview) | 61.43 | #99 |
| 11 | Gemma 4 31B | 59.21 | #382 |
| 12 | Claude Sonnet 5.5 | 59.04 | #24 |
| 13 | GPT-5.4 Mini | 57.6 | #406 |
| 14 | GPT-6 Luna | 55.98 | #105 |
| 15 | GPT-OSS-120B | 55.4 | #704 |
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-small · How It Works · Data refreshed daily, snapshot 2026-11-21.