DeepTumorVQA Agent (Vision Tools): 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 runs models as agents in the benchmark tool environment with only image-crop and medical knowledge tools; accuracy in percent, higher is better.
Metric: Answer accuracy (%) on the balanced 10,000-question multiple-choice test subset of DeepTumorVQA, the benchmark ReAct agent loop (up to 8 steps) over 2D inputs, vision mode: only organ-crop and medical knowledge tools, no quantitative measurements; higher is better. Source: arxiv.org. Saturation forecast: Around March 2027. 3 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3.5 9B | 40.3 | #838 |
| 2 | Gemini 3 Flash | 35.3 | #180 |
| 3 | Meissa-4B | 31.5 |
No result here: #2 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=deeptumorvqa-agent-vision-tools · How It Works · Data refreshed daily, snapshot 2026-11-22.