WearableQA - Overall: leaderboard
Metric: Accuracy (%; 10-option multiple choice, chain-of-thought). Source: arxiv.org. 14 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 72.9 |
| 2 | Claude Opus 4.6 | 60.2 |
| 3 | Claude Sonnet 4 | 53.2 |
| 4 | GPT-5.4 | 51.2 |
| 5 | Gemini 2.5 Pro | 50.6 |
| 6 | Gemma 4 26B A4B | 42.5 |
| 7 | GPT-4o | 34.7 |
| 8 | Mistral Small 3.1 | 29.3 |
| 9 | Llama 3.3 70B | 29.2 |
| 10 | Gemma 3 27B | 26.2 |
| 11 | Gemma 3 12B | 24.6 |
| 12 | Llama 3.1 8B | 23.8 |
| 13 | Gemma 3 4B | 19.8 |
| 14 | Llama 3.2 3B | 19.6 |
Interactive version: theaggregate.ai/benchmark?slug=wearableqa-overall · How It Works · Data refreshed daily, snapshot 2026-09-19.