OpenTumorBoard - Specialist Turn: leaderboard
From recorded multidisciplinary tumor-board meetings, the model plays a specialist and answers a question actually raised in the discussion, given the case summary and slides. An LLM judge scores clinical equivalence to the real specialist's answer on a 1-5 scale, higher is better (OpenTumorBoard, 2026).
Metric: Clinical equivalence (1-5) of the model's answer to the real specialist's answer, mean over 4,844 test questions from recorded tumor boards (model given the case summary and slides as image with caption, or caption only for text-only models; judged by Qwen3.8-27B); higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 9 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Ministral 3 14B | 3.02 | #1845 |
| 2 | Llama 4 Scout | 2.97 | #1776 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=opentumorboard-specialist-turn · How It Works · Data refreshed daily, snapshot 2026-11-13.