Is it possible to upload data files or the data itself, train a model on the data, save the model and then query it with new data using the API? Where is documentation on how to do that?
Hi and welcome to the Developer Forum!
You will need to look into embeddings, these allow you to chunk your data into blocks and then store them in a special kind of database called a vector database using the ada-002 model and once done you can search that for similarity to a search query, the system will then find data that is similar to the query and you can use that as context for an AI API call which will then combine the retrieved information and the users question and produce an answer from that.