AnnouncementPinecone serverless on AWS is now generally availableLearn more
Models

udever-bloom-7b1

Finetuned version of bigscience/bloom-7b1 that excels at multi-lingual text embeddings.
Dimension:Size of a single vector
supported by this model.
4096
Distance Metric:Used to measure similarity
between vectors.
cosine
Max Seq. Length:Number of tokens the model
can process at once.
2048

Overview

udever-bloom-7b1 is finetuned from bigscience/bloom-7b1 via BitFit on MS MARCO Passage Ranking, SNLI and MultiNLI data. It is a universal embedding model across tasks, natural and programming languages.

Udever stands for “Universal DEcoder VEctoR).”

This model is large (~28 gb), which might present difficulties for casual users.

This is the highest-performing model of the udever-bloom family of models.

Using the Model

Installation:

Creating Embeddings:

Learn more about udever-bloom-7b1