DDx-TRACE - Differential Diagnosis: leaderboard
Metric: Final four-item differential diagnosis score (0-1, a clinically plausible ranked set) 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) | 0.64 |
| 2 | Gemini 3 Flash | 0.61 |
| 3 | GPT-5.4 | 0.51 |
| 4 | Claude Sonnet 4.6 | 0.47 |
| 5 | GPT-5.4 Mini | 0.41 |
| 6 | Qwen 3.5 35B A3B | 0.37 |
| 7 | Qwen 3.5 27B | 0.35 |
| 8 | Qwen 3.5 9B | 0.28 |
| 9 | Gemma 3 27B | 0.25 |
| 10 | MedGemma-27B-IT | 0.24 |
| 11 | Qwen 3.5 4B | 0.2 |
| 12 | Lingshu-32B | 0.18 |
| 13 | MedGemma 1.5 4B | 0.11 |
Interactive version: theaggregate.ai/benchmark?slug=ddx-trace-differential-diagnosis · How It Works · Data refreshed daily, snapshot 2026-10-07.