A single-CPU/GPU lab exploring AI inference across LLMs, vision, audio, and multimodal models through serving, benchmarking, and optimization.
