Classify GPU's based on their benchmark score in order to provide an adaptive experience.
Advanced, lightweight system and OS information library
Run AI models locally on your machine with node.js bindings for llama.cpp. Enforce a JSON schema on the model output on the generation level