Vapi is the Voice AI platform for developers.

Vapi AI

Hello under voice configuration in the assistants dashboard, there is an area that lets you select a model (eleven English v1, turbo v2, turbo v2.5, and multilingual v2). It's been noticed that the assistant works best when we leave the model section blank. The question is, when we leave the model section blank, what is happening? Is it auto choosing one of the options without showing us? Is there a default option?

You can continue this conversation by mentioning me in the message.

<@425527778138390528>
Hello! When you leave the model section blank in the voice configuration of your assistant's dashboard, the system typically defaults to a standard or pre-configured model. This default setting is designed to provide a balance between performance and general applicability for a wide range of use cases.

The system may either be:
1. Automatically selecting the most suitable model based on the assistant's usage patterns and other configuration settings.
2. Using a generic model that is considered to be the best all-rounder for various tasks.

However, the exact behavior can depend on the specific implementation of the VAPI system. If you need more detailed information about what happens when the model section is left blank, please ask support for more info on this. They can provide you with the specifics of the default behavior for your assistant's voice configuration.

<@425527778138390528> can you share a loom video?

https://www.loom.com/share/362622f3a9824c13a1bab5824a60ef5a

<@425527778138390528> use turbo v2 model it will work best for you along with deepgram nova2-phonecall.

Thanks Shubham. Can you tell me which one it defaults to if we leave it on "select a model"? And why is turbo v2 the best and not turbo v2.5?

turbo_v2 is stable model compared to turbo_v2.5 and it's been around for long.

Thanks for you know the answer for the default model if we do not select anything?