superduper
superduper copied to clipboard
superduper-io
Superduper: Integrate AI models and machine learning workflows with your database to implement custom AI applications, without moving your data. Including streaming inference, scalable model hosting,...
**Tasks** - [ ] Migrate plugins to open-source `superduper` - [ ] Create template - [ ] Migrate streamlit - [ ] Make sure structure is as below Aim is...
We want to leverage rest API to build a simplified version on the Superduper interface. For this the rest API and a simple server implementation should go to OSS.
Currently this is the only part of the project which doesn't strictly follow the plugin structure. - [ ] lance - [ ] qdrant - [ ] "local"
By default, `.predict_batches` can call `.predict_many`. `.predict_many` will be the exact analogue of `.predict` in terms of signature.
This will make code much easier to read.
A function with a `.predict(x)` will be `"singleton"` whereas `.predict(x, y, z=1)` will be `*args, **kwargs`.
Metadata
Owner
Metadata
Superduper: Integrate AI models and machine learning workflows with your database to implement custom AI applications, without moving your data. Including streaming inference, scalable model hosting,...