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
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 94 |
| 2 | GPT-5.4 | 94 |
| 3 | Claude Sonnet 4.6 | 93.3 |
| 4 | Qwen 3.5 Plus | 91.3 |
| 5 | GPT-5.4 Mini | 89.3 |
| 6 | GLM-5 | 88.7 |
| 7 | Claude Haiku 4.5 | 88 |
| 8 | Qwen 3.5 397B A17B | 88 |
| 9 | DeepSeek V3.2 (Thinking) | 87.3 |
| 10 | Qwen 3.5 9B | 76.7 |
| 11 | Qwen 3.5 4B | 69.3 |
| 12 | MiniMax-M2.7 | 57.3 |
Interactive version: theaggregate.ai/benchmark?slug=smh-bench-interactive-agent-atomic-control · How It Works · Data refreshed daily, snapshot 2026-09-29.