SMH-Bench (Interactive Agent) - Atomic Control: leaderboard

Metric: Success rate (%) on Atomic Control (TC1, 150 single-intent device operations under clear, colloquial or noisy wording); Environment-Interactive Agent: a ReAct loop that starts from partial observability and queries and controls the HomeEnv simulator through tools under a call budget; final home state checked against the goal with untouched devices preserved; clarification and query answers judged by GPT-5; higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 13 models tracked.

Top models

#ModelScore
1Gemini 3.1 Pro (Preview)94
2GPT-5.494
3Claude Sonnet 4.693.3
4Qwen 3.5 Plus91.3
5GPT-5.4 Mini89.3
6GLM-588.7
7Claude Haiku 4.588
8Qwen 3.5 397B A17B88
9DeepSeek V3.2 (Thinking)87.3
10Qwen 3.5 9B76.7
11Qwen 3.5 4B69.3
12MiniMax-M2.757.3

Interactive version: theaggregate.ai/benchmark?slug=smh-bench-interactive-agent-atomic-control · How It Works · Data refreshed daily, snapshot 2026-09-29.