SMH-Bench (Interactive Agent) - Automation Scheduling: leaderboard

Metric: Success rate (%) on Automated Task Scheduling (TC4, 150 time- or state-triggered rules, trigger and action both verified); 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: Around December 2026. 13 models tracked.

Top models

#ModelScore
1Qwen 3.5 397B A17B69.3
2Claude Sonnet 4.668.7
3Gemini 3.1 Pro (Preview)64
4DeepSeek V3.2 (Thinking)64
5GPT-5.4 Mini50.7
6GPT-5.449.3
7Claude Haiku 4.548
8GLM-542.7
9Qwen 3.5 Plus42.7
10MiniMax-M2.742
11Qwen 3.5 9B36
12Qwen 3.5 4B21.3

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