DeepTumorVQA - Medical Reasoning: leaderboard

DeepTumorVQA (2026) is a hierarchical visual question answering benchmark on abdominal 3D CT that follows the evidence chain of tumor diagnosis, scored as multiple-choice accuracy on a balanced 10,000-question test subset. This board is accuracy in percent on the medical reasoning questions that grade or stage disease from measured values against clinical thresholds, answered directly from the image input without tools; higher is better.

Metric: Accuracy (%) on the Medical-reasoning-stage questions (two to four options: steatosis grade, T-stage, resectability from measured values against clinical thresholds) of the balanced 10,000-question multiple-choice test subset of DeepTumorVQA (42 subtypes over abdominal 3D CT volumes), direct inference zero-shot with each model's native input (2D organ-centred slices for 2D models, the full 3D volume for 3D models); higher is better. Source: arxiv.org. Saturation forecast: Around 2029. 12 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1LLaVA-OneVision-7B39.4#2981
2Qwen 3.5 35B A3B38.7#492
3Gemini 3 Flash33.9#175
4LLaVA-Med30.3
5MedGemma 1.5 4B29#2716
6Meissa-4B28.5
7HuatuoGPT-Vision-7B27.7#2162
8Qwen 3.5 9B27.4#828
9RadFM26.3
10Qwen 3 VL 4B25#1151
11InternVL3-8B23.1#1782
12Qwen 3 VL 8B21.8#917

No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=deeptumorvqa-medical-reasoning · How It Works · Data refreshed daily, snapshot 2026-11-21.