What is the best way to create an assistant that uses a large amount of data?

I want to build an AI assitant that will answer users question but the thing is I want this assistant to be based on a dataset that changes over times likes prices etc. So my question is how to fine tune a model with this or to create an embeling model for this. Thx

If your dataset is constantly changing, like prices, using embeddings as an external source for your model would be the best option.

1 Like

So I need to use embedding and Fine-tunning to achieve this model and if so how to combine both.

Fine-tuning is not used to “educate” the AI on your content. For that, use embeddings only.