VAPI•2y ago

Finetune Model (e.g. llama3)

Has anyone tried finetuning an open source model for conversation flow? To make the agent use industry-specific vocab.

I'm looking for a finetuned model that uses a more casual tone when completing replies.

Ask Inkeep•5/23/24, 4:13 PM

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

Ask Inkeep•5/23/24, 4:13 PM

-----------------------------
While I generate an answer, here are some sources that might be helpful:

Connecting Your Custom LLM to Vapi: A Comprehensive Guide
Custom LLM
Orchestration Models
Orchestration Models
-----------------------------

Ask Inkeep•5/23/24, 4:13 PM

Answer

Vapi supports using any OpenAI-compatible endpoint as the Large Language Model (LLM), including the possibility of using fine-tuned or custom models that are tailored to specific vocabularies or industries. This enables the integration of models like Llama3, provided they are compatible with OpenAI's API standards. You can set up these models by specifying your custom LLM provider and model details in the configuration of your assistants in Vapi (1) .

Sources
[1] Custom LLM