<h2>llama-cpp-turboquant Overview</h2><p><strong>llama-cpp-turboquant</strong> (TheTom/llama-cpp-turboquant) is a production-focused fork of <strong>llama.cpp</strong> that implements <strong>TurboQuant</strong> (Google Research, arXiv:2504.19874, ICLR 2026) — extreme low-bit quantization of the KV cache (and model weights) using PolarQuant + Walsh-Hadamard rotation + Lloyd-Max optimal scalar quantization. It compresses the KV cache 3.8x-6.4x