Vista (/vi’sta/) is our tool to visualize embeddings in three dimensions:


It can be used to understand question/answer, documents used in knowledge retrieval, to compare models to each other, and to annotate a set of data.

The python interface is easy to use, simply import the model and then call it on a set of:

import pykoi as pk
# files
files = pk.UploadFiles('./path/to/files')

# visualize files with vista
viz =

# launch app
app = pk.App()

This will give you a sharable interface with the embedding model: