DeepSeek-V3, GPT-4, Phi-4, and LLaMA-3.3 generate correct code for LoRaWAN-related engineering tasks

TL;DR

This study evaluates 16 large language models for automating LoRaWAN engineering tasks, finding that some lightweight models perform comparably to advanced models like GPT-4 and DeepSeek-V3 in generating correct Python code.

Contribution

It demonstrates that certain smaller LLMs can effectively perform complex engineering tasks, expanding options for lightweight, local deployment.

Findings

DeepSeek-V3 and GPT-4 consistently accurate

Some lightweight models like Phi-4 and LLaMA-3.3 perform well

Model performance depends on prompt design and fine-tuning

Abstract

This paper investigates the performance of 16 Large Language Models (LLMs) in automating LoRaWAN-related engineering tasks involving optimal placement of drones and received power calculation under progressively complex zero-shot, natural language prompts. The primary research question is whether lightweight, locally executed LLMs can generate correct Python code for these tasks. To assess this, we compared locally run models against state-of-the-art alternatives, such as GPT-4 and DeepSeek-V3, which served as reference points. By extracting and executing the Python functions generated by each model, we evaluated their outputs on a zero-to-five scale. Results show that while DeepSeek-V3 and GPT-4 consistently provided accurate solutions, certain smaller models -- particularly Phi-4 and LLaMA-3.3 -- also demonstrated strong performance, underscoring the viability of lightweight alternatives. Other models exhibited errors stemming from incomplete understanding or syntactic issues. These findings illustrate the potential of LLM-based approaches for specialized engineering applications while highlighting the need for careful model selection, rigorous prompt design, and targeted domain fine-tuning to achieve reliable outcomes.