PersonalHomeBench - Information Grounding (Sole Reasoning): leaderboard

PersonalHomeBench (2026) tests models as assistants in generated, persona-rich smart homes. Information grounding asks factual and multi-hop questions about the household; accuracy in percent, with the full household context in the prompt and no tools. Higher is better.

Metric: Accuracy (%) on 1,000 information grounding questions (factual lookup to multi-hop reasoning over household entities; exact match, else a Gemini 2.5 Pro semantic-equivalence judge), over the text-based split of PersonalHomeBench (1,100 generated, manually reviewed households with persona-based occupants and appliances), temperature 0, in the sole-reasoning setting (no tools; the full household context, occupant profiles and memories are given in the prompt); each model at its stated reasoning mode; higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 7 models tracked. Updated 7 Oct 2026.

Top models

#ModelSettingScoreOverall rank
1GPT-OSS-20BHigh81.2#1183
2GPT-4o80#751
3Qwen 3 30BThinking78.82
4GPT-477.9#970
5Qwen 3 4BReasoning75.7#2348
6nemotron-3-nano-30B-a3B75.62#1946
7Qwen 3 30B Instruct73.5#1365

No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=personalhomebench-information-grounding-sole-reasoning · How It Works · Data refreshed daily, snapshot 2026-11-21.