UMLoad: Fast On-Demand Model Loading on Unified Memory Architecture

arXiv:2611.04253 · cs.OS, cs.AR · Submitted 2026-11-03 · Read on arXiv

cs.OS, cs.AR

Submitted: 2026-11-03

Updated: 2026-11-04

Code: https://github.com/ggml-org/llama.cpp

Terminology

Sources

Related papers