Edge-LLM-Inference-Benchmark

Measurement study of LLM inference trade-offs across edge-continuum hardware, including accuracy, latency, model footprint, energy, and streaming-delay sensitivity analysis.