LocalBox

LocalBox runs GGUF language models locally through a bundled llama.cpp CPU
runtime. Download a model, chat in your browser, and use the local API.
No dedicated GPU is required. Model memory requirements apply; the Lite model
needs at least 4 GiB RAM, with 8 GiB recommended. Internet access is needed
for model downloads. Published by Menayer; created by Amr El Menayer.