Small language models are agentic but not autonomous in greenhouse climate control
Agentic large language models (LLMs) are proposed for cyber-physical control, yet evidence on how reliably small open-weight models turn sensor context into actuator commands is scarce. We present a reproducible benchmark grounded in an operational ESP32-based greenhouse controller: ninety-six generated test cases run against the production tool schema and prompts, scored deterministically, with the controller’s rule-based fallback ported verbatim as baseline. Four quantized models (3–14 B) ran on a pinned llama.cpp stack in two locked environments two months apart (12,000 scored runs). Three findings dominate. Autonomous self-initiation is far less reliable than command execution: models executing 65–100% of chat commands fall to 0.4–64.6% when acting unprompted. Model rankings are environment-conditional: one environment favoured Qwen2.5-7B over the 14B (McNemar p = 0.016; cluster-robust GEE p = 0.084), the other decisively reversed it at identical seeds ( p < 0.001; 77.5% cross-environment agreement), requiring re-benchmarking on the deployment stack. The rule baseline matches the second-best model with the best refusal behaviour: the LLM’s measured value is language-mediated actuation, knowledge capture, and auditability, not superior regulation. Field telemetry and thermal replay show the hybrid controller running unattended through the growing season, rules holding the safety-critical floor. All instruments and both environments’ traces are released.
Authors
- Mihai Penica (ORCID: https://orcid.org/0000-0003-1050-7617)
- M. Bhattacharya
- E. O’Connell
- A. Dooley
Institutions
- University of Limerick (IE)
- Illinois State University (US)
Publication Details
- Journal
- Scientific Reports
- Published
- 2026-11-17
- DOI
- https://doi.org/10.1038/s41598-026-78255-1
- Primary Topic
- Smart Agriculture and AI
- Type
- article
- Field-Weighted Citation Impact
- 0.00