Edge-Deployable LLM Fine-Tuning on a Single GPU for Telecom Network Troubleshooting
For telecom operators needing to deploy LLMs at edge sites with limited GPU resources, this work offers practical profiling insights and configuration guidelines, though it is incremental in nature.
This paper profiles LLM fine-tuning on a single GPU for telecom edge troubleshooting, showing that reasoning and non-reasoning models behave differently during SFT and RFT due to chat template and control flag differences. It provides configuration guidelines for stable, efficient fine-tuning under edge constraints.
Telecom troubleshooting at edge sites requires low-latency model responses and localized model adaptation to satisfy operational and data sovereignty requirements. However, deploying large language models (LLMs) at telecom edge sites is constrained by limited power, cooling, space, and weight budgets for GPU infrastructure. These challenges are further amplified by human-patterned Radio Access Network (RAN) traffic that often results in low GPU utilization and poor return on investment, as well as by architectural mismatches between deterministic ASIC-based telecom processing and GPU-oriented AI workloads. Consequently, single-GPU fine-tuning becomes a practical requirement for scalable edge AI deployment rather than merely a resource limitation. This paper presents a GPU profiling study of LLM fine-tuning using the Unsloth framework on a single edge-class accelerator. We systematically analyze the effects of maximum sequence length, GPU memory utilization, Low-Rank Adaptation (LoRA) rank, and generation count on training stability and resource efficiency. We further investigate trade-offs in KV cache usage, activation memory overhead, and runtime stability under inductor compilation. In addition, we show that reasoning and non-reasoning model architectures exhibit substantially different behaviors during supervised fine-tuning (SFT) and reinforcement fine-tuning (RFT) because of differences in chat template structures, reasoning tags, and control flags. Experiments are conducted on a telecom troubleshooting dataset consisting of question-answer pairs augmented with top-3 retrieved contextual documents. The results provide practical configuration guidelines for stable, efficient, and resource-aware LLM fine-tuning in telecom edge environments.