Add Text Embedding Feature in API Call by Integrating Self-Hosted Endpoint Server

[elohffa]

Description

Currently, our API does not support text embedding, which can limit the functionality and flexibility of our application, especially for functionalities that involves natural language processing or machine learning.

Why

I propose that we add a text embedding feature to our API calls by integrating a self-hosted endpoint server. This will enable our API to convert input text data into meaningful numerical representations (embeddings) that can be used for further processing, analysis or modeling.

Integrating text embedding in our API will allow us to harness the power of advanced techniques in machine learning and natural language processing. This can significantly improve the capabilities of our application, enabling it to handle complex tasks such as text classification, semantic analysis, and more.

Alternatives

Modify our current API that accept openai embed api call in our call and then include a new option to relay this url to user self define url , This endpoint will accept raw text data, pass them to the self-hosted server for processing, and then return the computed embeddings to the calling client.