PEC-Home (RAG) - Multi-User Level 2: leaderboard

Metric: Execution accuracy (%; 1,068 smart-home commands at ellipsis level 2 of 4 (progressively elliptical), functionally correct device operations; dialogue history retrieved from a vector store, top-3). Source: arxiv.org. Saturation forecast: Estimated already saturated. 10 models tracked.

Top models

#ModelScore
1DeepSeek V392.88
2GPT-4o92.7
3Qwen 2.5 72B Instruct92.32
4Qwen 2.5 7B Instruct89.79
5Qwen 2.5 32B Instruct87.92
6Gemma 2 27B (IT)87.36
7Qwen 2.5 14B Instruct86.61
8Mistral 7B Instruct (v0.3)83.15
9Gemma 2 9B (IT)82.68
10Llama 3.1 8B Instruct71.82

Interactive version: theaggregate.ai/benchmark?slug=pec-home-rag-multi-user-level-2 · How It Works · Data refreshed daily, snapshot 2026-09-26.