We want users to be able to generate embeddings on the server and not have to bring their own. We'll need to create a new EmbeddingFunction class than can be extended for custom embedding models on the server, custom endpoints, as well as OpenAI, Cohere, Sentence Transformers, MiniLM, etc.
We want users to be able to generate embeddings on the server and not have to bring their own. We'll need to create a new EmbeddingFunction class than can be extended for custom embedding models on the server, custom endpoints, as well as OpenAI, Cohere, Sentence Transformers, MiniLM, etc.