MedFlowBench (Radiology) - Brain MRI Strict: leaderboard
Metric: Strict score (0-1, shown times 100) on UCSF-PDGM multi-sequence brain MRI (495 cases: tumor diagnosis): the diagnosis must be right and the reported axial, sagittal and coronal key slices must fall within the hidden tumor mask; 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 January 2027. 14 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 73 | #54 |
| 2 | GPT-5.5 | 66 | #26 |
| 3 | Gemini 3 Flash (Preview) | 59 | #78 |
| 4 | Qwen 3.5 27B | 27 | #183 |
| 5 | Claude Sonnet 4.6 | 18 | #85 |
| 6 | GPT-5.4 Mini | 18 | #202 |
| 7 | Qwen 3.5 35B A3B | 15 | #250 |
| 8 | Gemma 3 27B (IT) | 11 | #509 |
| 9 | Claude Opus 4.7 | 11 | #45 |
| 10 | MedGemma 1.5 4B | 9 | #775 |
| 11 | Qwen 3.5 4B | 6 | #470 |
| 12 | Qwen 3.5 9B | 5 | #363 |
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-brain-mri-strict · How It Works · Data refreshed daily, snapshot 2026-10-11.