Vapi is the Voice AI platform for developers.

Vapi AI

What is the granularity of the RAG in terms of how many document excerpts are provided? Is it variable or dependent on embedding match strength? How large are the excerpts? Are any of those parameters adjustable?

Hi <@209031094631137280>, I'm an AI assistant for Vapi. I'll analyze available sources and get back to you shortly.

-----------------------------
**While I generate an answer, here are some sources that might be helpful**:
* [Launch Screen Assets](<https://github.com/VapiAI/flutter/blob/main/example/ios/Runner/Assets.xcassets/LaunchImage.imageset/README.md>)
* [Vapi Python SDK](<https://github.com/VapiAI/python/blob/main/README.md>)
* [Vapi Web SDK](<https://github.com/VapiAI/web/blob/main/README.md>)
* [Vapi iOS SDK](<https://github.com/VapiAI/ios/blob/main/README.md>)
-----------------------------

### Answer
I'm sorry, but the information sources available to me do not provide specific details about the granularity of the Retrieval-Augmented Generation (RAG) in terms of document excerpts, their dependency on embedding match strength, excerpt sizes, or adjustability of these parameters in the context of Vapi. For more specialized inquiries like this, I recommend consulting additional documentation or resources specific to the technology you are using.

Got it. Tagging <@&1231325148079194222> for additional help.

Can you please ask this question on Vapi OH hour?

You can check the events section it most probably tomorrow.