DDx-TRACE - Final Diagnosis: leaderboard
Metric: Final top-1 diagnosis score (0-1, rubric-graded) on DDx-TRACE (211 physician-adjudicated multimodal neuroradiology cases), the model starts from limited history, requests imaging in free form (at most 6 requests), updates a four-item probabilistic differential after each turn and stops with a localized diagnosis, provider default decoding; endpoints judged against case rubrics by Gemini 3 Flash at temperature 0; higher is better. Source: arxiv.org. Saturation forecast: Around January 2027. 13 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 0.55 |
| 2 | Gemini 3 Flash | 0.54 |
| 3 | GPT-5.4 | 0.43 |
| 4 | Claude Sonnet 4.6 | 0.42 |
| 5 | GPT-5.4 Mini | 0.35 |
| 6 | Qwen 3.5 27B | 0.3 |
| 7 | Qwen 3.5 35B A3B | 0.29 |
| 8 | Qwen 3.5 9B | 0.21 |
| 9 | Gemma 3 27B | 0.19 |
| 10 | MedGemma-27B-IT | 0.17 |
| 11 | Qwen 3.5 4B | 0.16 |
| 12 | Lingshu-32B | 0.15 |
| 13 | MedGemma 1.5 4B | 0.08 |
Interactive version: theaggregate.ai/benchmark?slug=ddx-trace-final-diagnosis · How It Works · Data refreshed daily, snapshot 2026-10-07.