The extension currently relies on OpenAI for embeddings of the text pieces.
There are two alternatives we could provide:
- other services
- local models
Drawback of the local models is they'd require more local compute power. But on the positive side - there would be less steps to setup the extension.
The extension currently relies on OpenAI for embeddings of the text pieces.
There are two alternatives we could provide:
Drawback of the local models is they'd require more local compute power. But on the positive side - there would be less steps to setup the extension.