This research explores how lightweight Large Language Models can provide effective malware detection directly on edge devices with limited computational resources.

• Evaluates performance of compact LLMs for identifying sophisticated malware at the network edge
• Addresses the challenge of implementing robust security in resource-constrained environments
• Demonstrates viable alternatives to resource-intensive cloud-based security solutions
• Offers practical architectural considerations for real-world deployment

This work is significant for cybersecurity professionals as it enables more responsive threat detection without the latency of cloud processing, potentially improving protection for IoT and edge computing environments where traditional security measures are challenging to implement.

Malware Detection at the Edge with Lightweight LLMs: A Performance Evaluation