Currently embedding_model, in library and in server, is the string text-embedding-ada-002.
But in the future there might be multiple models in production. Should we prepend a namespace in front of each? E.g. openai.com:text-embedding-ada-002.
If we expect there to be a small number of hard-coded models supported, then we can leave it as is--we can create a page of documentation about which each one is.
But if we expect there to be a large number and a number that increases even when version of this library doesn't increment, then we should probably include a namespace.
dglazkov
commented
1 year ago
Currently
embedding_model, in library and in server, is the stringtext-embedding-ada-002.But in the future there might be multiple models in production. Should we prepend a namespace in front of each? E.g.
openai.com:text-embedding-ada-002.If we expect there to be a small number of hard-coded models supported, then we can leave it as is--we can create a page of documentation about which each one is.
But if we expect there to be a large number and a number that increases even when
versionof this library doesn't increment, then we should probably include a namespace.