WearableQA - Overall (Direct Answer): leaderboard
Metric: Accuracy (%; 10-option multiple choice, direct answer). Source: arxiv.org. 14 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 71.3 |
| 2 | Claude Sonnet 4 | 48.3 |
| 3 | Claude Opus 4.6 | 40.6 |
| 4 | Gemma 4 26B A4B | 35.7 |
| 5 | GPT-5.4 | 33.5 |
| 6 | Gemini 2.5 Pro | 32 |
| 7 | Llama 3.3 70B | 27.2 |
| 8 | Mistral Small 3.1 | 24.5 |
| 9 | GPT-4o | 23.9 |
| 10 | Gemma 3 27B | 23.8 |
| 11 | Llama 3.1 8B | 22.3 |
| 12 | Gemma 3 12B | 21.7 |
| 13 | Gemma 3 4B | 20.6 |
| 14 | Llama 3.2 3B | 17.9 |
Interactive version: theaggregate.ai/benchmark?slug=wearableqa-overall-direct-answer · How It Works · Data refreshed daily, snapshot 2026-09-19.