mlx-bun library API
    Preparing search index...
    interface FuseDependencies {
        catalog: Pick<ModelCatalog, "find" | "canPublish" | "publish">;
        fuse: (
            modelDir: string,
            adapterDir: string,
            outDir: string,
            onProgress?: (e: FuseProgress) => void,
            options?: FuseOptions,
        ) => Promise<FuseStats>;
        terminal: CliTerminal;
        exists(path: string): boolean;
        log(line: string): void;
        modelsDir(): string;
    }
    Index
    catalog: Pick<ModelCatalog, "find" | "canPublish" | "publish">

    Resolves a query the way the catalog does (a model directory as given, else a downloaded model); the write token and the push behind --upload-repo are the catalog's too.

    fuse: (
        modelDir: string,
        adapterDir: string,
        outDir: string,
        onProgress?: (e: FuseProgress) => void,
        options?: FuseOptions,
    ) => Promise<FuseStats>

    Type Declaration

      • (
            modelDir: string,
            adapterDir: string,
            outDir: string,
            onProgress?: (e: FuseProgress) => void,
            options?: FuseOptions,
        ): Promise<FuseStats>
      • Fuse the adapter at adapterDir into the base model at modelDir and write a complete, loadable snapshot to outDir (weights + config.json + tokenizer aux files). Quantization layout is preserved verbatim: fused modules are dequantized, updated, and re-quantized with their own source spec; untouched modules pass through bit-identical. With options.dequantize every quantized module is written dense instead.

        Parameters

        • modelDir: string
        • adapterDir: string
        • outDir: string
        • OptionalonProgress: (e: FuseProgress) => void
        • options: FuseOptions = {}

        Returns Promise<FuseStats>

    terminal: CliTerminal