DementiaCare-Bench - 64 Frames: leaderboard
Metric: Accuracy (%). Source: arxiv.org. Saturation forecast: Around December 2026. 12 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemma 3 27B (IT) | 85.1 |
| 2 | Claude Sonnet 4.6 | 83.2 |
| 3 | Gemini 2.5 Flash | 81.4 |
| 4 | Qwen 3 VL 8B | 78.6 |
| 5 | GPT-4o | 78.3 |
| 6 | GLM-4.6V | 71.2 |
| 7 | InternVL3-8B | 65.8 |
Interactive version: theaggregate.ai/benchmark?slug=dementiacare-bench-64-frames · How It Works · Data refreshed daily, snapshot 2026-09-24.