PEC-Home (ICL) - Dynamic Preferences Level 1: leaderboard

Metric: Execution accuracy (%; 712 smart-home commands at ellipsis level 1 of 4 (complete commands), functionally correct device operations; one in-context example per ellipsis level, no dialogue history). Source: arxiv.org. Saturation forecast: Estimated already saturated. 10 models tracked.

Top models

#ModelScore
1GPT-4o93.26
2DeepSeek V392.84
3Gemma 2 9B (IT)91.85
4Gemma 2 27B (IT)91.01
5Qwen 2.5 72B Instruct88.9
6Mistral 7B Instruct (v0.3)88.76
7Qwen 2.5 32B Instruct88.2
8Qwen 2.5 7B Instruct86.94
9Qwen 2.5 14B Instruct81.88
10Llama 3.1 8B Instruct74.58

Interactive version: theaggregate.ai/benchmark?slug=pec-home-icl-dynamic-preferences-level-1 · How It Works · Data refreshed daily, snapshot 2026-09-26.