aisdk.slm

Local Small Language Model Inference

A self-contained engine for running local small language models: quantized GGUF models and ONNX/torch models on local hardware, plus model discovery and weight downloading. Designed to plug into the aisdk ecosystem while remaining usable on its own.

• Status: development (GitHub)
• Source: github.com/YuLab-SMU/aisdk.slm

Install

# install.packages("pak")
pak::pak("YuLab-SMU/aisdk.slm")

Highlights

• Run quantized GGUF models locally.
• ONNX / torch model backends.
• Model discovery and downloading into a local cache.