Energy Efficiency of Locally Deployed LLMs: A Preliminary Quantitative GPU Power Benchmark on Consumer Hardware
Original ↗

Energy Efficiency of Locally Deployed LLMs: A Preliminary Quantitative GPU Power Benchmark on Consumer Hardware

AI Research arxiv.org · 12h ago· Philipp M. Z\"ahl, Anika Hennig Read on arxiv.org ↗
0:00 / 0:00