https://github.com/mindtro/semafold

Vector compression with TurboQuant codecs for embeddings, retrieval, and KV-cache. 10x compression, pure NumPy core — optional GPU acceleration via PyTorch (CUDA/MPS) or MLX (Metal).

embedding-compression kv-cache llm-inference qjl quantization retrieval semafold turboquant vector-compression vector-database

Last synced: 07 Apr 2026