PEC-Home (Zero-Shot) - Dynamic Preferences Level 1: leaderboard
Metric: Execution accuracy (%; 712 smart-home commands at ellipsis level 1 of 4 (complete commands), functionally correct device operations; zero-shot prompt with the home state, no dialogue history). Source: arxiv.org. Saturation forecast: Estimated already saturated. 10 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Mistral 7B Instruct (v0.3) | 57.16 |
| 2 | Qwen 2.5 7B Instruct | 54.21 |
| 3 | GPT-4o | 52.81 |
| 4 | DeepSeek V3 | 52.81 |
| 5 | Qwen 2.5 32B Instruct | 50.14 |
| 6 | Gemma 2 27B (IT) | 45.51 |
| 7 | Gemma 2 9B (IT) | 44.94 |
| 8 | Qwen 2.5 14B Instruct | 38.48 |
| 9 | Llama 3.1 8B Instruct | 36.66 |
| 10 | Qwen 2.5 72B Instruct | 34.27 |
Interactive version: theaggregate.ai/benchmark?slug=pec-home-zero-shot-dynamic-preferences-level-1 · How It Works · Data refreshed daily, snapshot 2026-09-26.