DDx-TRACE - Trajectory Confidence: leaderboard
DDx-TRACE (2026) scores diagnostic trajectories on physician-adjudicated neuroradiology cases: models request imaging in free form, update a probabilistic differential diagnosis after each turn and stop with a localized diagnosis, judged against case rubrics. This board scores the same confidence alignment averaged over all turns, from -1 to 1; higher is better.
Metric: Trajectory confidence (-1 to 1): the turn-averaged confidence alignment of the evolving differential 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 February 2027. 13 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 0.501 | #99 |
| 2 | Gemini 3 Flash | 0.498 | #175 |
| 3 | GPT-5.4 | 0.407 | #135 |
| 4 | Claude Sonnet 4.6 | 0.396 | #173 |
| 5 | GPT-5.4 Mini | 0.334 | #406 |
| 6 | Qwen 3.5 27B | 0.285 | #375 |
| 7 | Qwen 3.5 35B A3B | 0.276 | #492 |
| 8 | Qwen 3.5 9B | 0.191 | #828 |
| 9 | Gemma 3 27B | 0.19 | #1663 |
| 10 | MedGemma-27B-IT | 0.163 | #1350 |
| 11 | Qwen 3.5 4B | 0.152 | #1241 |
| 12 | Lingshu-32B | 0.148 | #1113 |
| 13 | MedGemma 1.5 4B | 0.08 | #2716 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=ddx-trace-trajectory-confidence · How It Works · Data refreshed daily, snapshot 2026-11-21.