MedFlowBench (Radiology) - Lung PET/CT Task: leaderboard
Metric: Task score (0-1, shown times 100) on NSCLC lung PET/CT (162 cases: tumor location, pathological T and N stage, histology and grade): the share of the five structured labels answered correctly, averaged over cases; the model drives 3D Slicer through the MedOpenClaw runtime with viewer-native actions only (series selection, scrolling, windowing, prealigned fusion), inspecting the complete study within a 20-round budget and returning an answer with structured evidence (Track A); higher is better. Source: arxiv.org. Saturation forecast: Around November 2027. 14 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.5 | 55 | #26 |
| 2 | Gemini 3.1 Pro (Preview) | 50 | #54 |
| 3 | Gemini 3 Flash (Preview) | 45 | #78 |
| 4 | Claude Opus 4.7 | 43 | #45 |
| 5 | Qwen 3.5 27B | 37 | #183 |
| 6 | Gemma 3 27B (IT) | 35 | #509 |
| 7 | Qwen 3.5 35B A3B | 35 | #250 |
| 8 | GPT-5.4 Mini | 31 | #202 |
| 9 | Claude Sonnet 4.6 | 26 | #85 |
| 10 | Qwen 3.5 4B | 20 | #470 |
| 11 | Qwen 3.5 9B | 16 | #363 |
| 12 | MedGemma 1.5 4B | 16 | #775 |
No result here: #3 Claude Opus 5.5, #5 GPT-6 Astra, #8 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=medflowbench-radiology-lung-pet-ct-task · How It Works · Data refreshed daily, snapshot 2026-10-11.