How to change the GPT model on the on-premise version

NOTE: This article is for the on-premise version. The SaaS version uses the latest model (GPT-5 at the time of writing)

If you're using the on-premise version of GPT, you can change the default model from GPT-5 to another model, for example GPT-4o, by following these simple steps:

Steps to Change the Model

1. Access the Configuration File:
   • Open the appsettings.json file in your application’s configuration directory.
2. Update the Model Name:
   • Look for the OpenAIModelName key. If it exists, update it; if not, add it. For  example:

"OpenAIModelName": "gpt-4o"

3. Save and Apply Changes:
   • Save the appsettings.json file.
   • Restart your application or service to apply the changes.

Additional Resources

For a full list of available models and further documentation, visit the OpenAI Models Documentation.

By modifying the OpenAIModelName in your appsettings.json file, you can easily switch to a different language model, enabling you to take advantage of the latest advancements in AI technology.

Make sure that your API key has access to the model that you choose. Also keep in mind that some models are significantly more expensive than others.