When uploading documents, I got an error that the vector ID was not ASCII. Is there plan to expand this to support UTF-8 strings?
I had the same problem, it turns out that it was because some of my documents had “ñ”, I just replaced those letters and that’s it
I doubt this would be something that they would roadmap. metadata fields are UTF-8 supported. The vector id field should be a UUID/GUID. Anything else that seems like a collision would be highly likely. You can always grab vectors by a single metadata key to get the id and CRUD it.
What is your use case for requiring UTF-8 in the vector id field?