quiad/llama/usedBin.json

Run AI models locally on your machine with node.js bindings for llama.cpp. Force a JSON schema on the model output on the generation level

{
    "use": "prebuiltBinaries"
}