Skip to main content
Version: 3.6.1

Manual Installation of AI Model Weights

If the automatic download of AI model weights failsβ€”due to internet restrictions or other issuesβ€”you can manually download and install the necessary files. This guide explains where to place the files and how to create the required directories if they do not already exist.

tip

First, check config.json to confirm which models are currently configured before downloading files. See the Check config.json section below.


AI Models Directory Structure​

Below is the directory structure of VRCT:

C:\Users\{YourUsername}\AppData\Local\VRCT
(or wherever the folder containing VRCT.exe is located)
β”œβ”€β”€ VRCT.exe
...
└── weights
β”œβ”€β”€ ctranslate2
β”‚ └── {CTRANSLATE2_WEIGHT_TYPE} ← folder name must match config.json exactly
β”‚ β”œβ”€β”€ config.json
β”‚ β”œβ”€β”€ model.bin
β”‚ └── ...
└── whisper
β”œβ”€β”€ tiny
β”œβ”€β”€ base (Required by default)
β”‚ β”œβ”€β”€ config.json
β”‚ β”œβ”€β”€ model.bin
β”‚ β”œβ”€β”€ tokenizer.json
β”‚ └── vocabulary.txt
β”œβ”€β”€ small
β”œβ”€β”€ medium
β”œβ”€β”€ large-v1
β”œβ”€β”€ large-v2
└── large-v3
important

The CTranslate2 folder name must exactly match the value of CTRANSLATE2_WEIGHT_TYPE in your config.json. For example, if config.json contains "CTRANSLATE2_WEIGHT_TYPE": "nllb-200-distilled-600M-ct2-int8", the folder must be named nllb-200-distilled-600M-ct2-int8.


Check config.json​

Before downloading anything, open your config.json and confirm these two values:

"CTRANSLATE2_WEIGHT_TYPE": "nllb-200-distilled-600M-ct2-int8",
"WHISPER_WEIGHT_TYPE": "base",
  • CTRANSLATE2_WEIGHT_TYPE β€” determines the CTranslate2 (translation) model to use and must match the folder name under weights\ctranslate2\.
  • WHISPER_WEIGHT_TYPE β€” determines the Whisper (transcription) model to use and must match the folder name under weights\whisper\.

Make sure you download and place the files that match these values.


Model Download URLs​

VRCT needs a CTranslate2 (translation) model and a Whisper (transcription) model.

CTranslate2 (Translation)​

The default translation model is nllb-200-distilled-600M-ct2-int8. Place all downloaded files into weights\ctranslate2\{CTRANSLATE2_WEIGHT_TYPE}\.

warning

Clicking the direct download links below will immediately start a download.

NLLB-200 Distilled 600M (default) β€” place in weights\ctranslate2\nllb-200-distilled-600M-ct2-int8\

Alternatively, browse the JustFrederik/nllb-200-distilled-600M-ct2-int8 model page on Hugging Face.


Whisper (Transcription)​

Download the specific Whisper model that matches your WHISPER_WEIGHT_TYPE setting. The base model is required by default.

Place all downloaded files into weights\whisper\{WHISPER_WEIGHT_TYPE}\.

ModelLinkNotes
tinyHugging Face
baseHugging Face(Required by default)
smallHugging Face
mediumHugging Face
large-v1Hugging Face
large-v2Hugging Face
large-v3Hugging Face

For each Whisper model, download these 4 files:

  • config.json
  • model.bin
  • tokenizer.json
  • vocabulary.txt
note

For Whisper models from the Systran series, only the above 4 files are needed. preprocessor_config.json and vocabulary.json are not required.

Direct download links for the base model:


Still having issues?​

  1. Check the folder name β€” Make sure the CTranslate2 folder name under weights\ctranslate2\ exactly matches CTRANSLATE2_WEIGHT_TYPE in config.json. A mismatch is the most common cause of failure.

  2. Check VPN / Proxy / Antivirus β€” If automatic download is failing, verify that your VPN, proxy, or antivirus software is not blocking connections to huggingface.co.

  3. Delete and re-download β€” Delete the weights\whisper and weights\ctranslate2 folders and let VRCT re-download them on the next launch.

  4. Attach logs when reporting β€” If the issue persists, attach your logs and settings data when filing a GitHub issue.