<h2>AirLLM Overview</h2><p><strong>AirLLM</strong> is an open-source Python library by <strong>Gavin Li</strong> that optimizes large language model inference memory usage, enabling 70B models on a single 4GB GPU and Llama 3.1 405B on 8GB VRAM without quantization, distillation, or pruning.</p><ul><li><span class="feature">Low-VRAM 70B Inference:</span> Run 70B-class LLMs on a single 4GB GPU card via layer-wise loa