Fails to download the large model

[binarykitchen]

It fails to auto-download the large model:

[dev:server] DEBUG: [Nodejs-whisper] Checking and downloading model if needed: large
[dev:server] DEBUG: autoDownloadModelName
[dev:server] DEBUG: options
[dev:server] DEBUG: [Nodejs-whisper] Auto-download Model: large
[dev:server] ERROR: [Nodejs-whisper] Error caught in autoDownloadModel:
[dev:server] ERROR: [Nodejs-whisper] Error during processing: [Nodejs-whisper] Failed to download model: 
[dev:server] ERROR: Operation failed: [Nodejs-whisper] Failed to download model: 
[dev:server]     err: {
[dev:server]       "type": "Error",
[dev:server]       "message": "Operation failed: [Nodejs-whisper] Failed to download model: ",
[dev:server]       "stack":
[dev:server]           Error: Operation failed: [Nodejs-whisper] Failed to download model: 
[dev:server]               at /home/michael-heuberger/code/binarykitchen/videomail.io/node_modules/nodejs-whisper/dist/index.js:50:19
[dev:server]               at Generator.throw (<anonymous>)
[dev:server]               at rejected (/home/michael-heuberger/code/binarykitchen/videomail.io/node_modules/nodejs-whisper/dist/index.js:6:65)
[dev:server]     }

Configuration is

    // When changing model, have to reinstall all
    await nodewhisper(reducedWav, {
      modelName: "large",
      // Like that, we don't have to type `npx nodejs-whisper download` every time.
      autoDownloadModelName: "large",
      logger: consoleLogger,
      removeWavFileAfterTranscription: true,
      // withCuda: true, // (optional) use cuda for faster processing
      whisperOptions: {
        outputInVtt: true,
        // Default is 20 which is too long
        timestamps_length: 14,
      },
    });

Using the latest version 0.2.9 here.

Any ideas why it fails for the large model but works fine for the other one, base.en?