On Epsilla Cloud, you can enable Mixedbread AI integration by providing your Mixedbread AI API key (we securely manage your keys using AWS KMS):
Epsilla integrates with Mixedbread AI with the following embedding models.
| Name | Dimensions |
|---|---|
| mixedbreadai/UAE-Large-V1 | 1024 |
| mixedbreadai/bge-large-en-v1.5 | 1024 |
| mixedbreadai/gte-large | 1024 |
| mixedbreadai/e5-large-v2 | 1024 |
| mixedbreadai/multilingual-e5-large | 1024 |
| mixedbreadai/multilingual-e5-base | 768 |
| mixedbreadai/gte-large-zh | 1024 |
For Epsilla open source vector db, you just need to add a header in the data ingestion and semantic search queries like this.
Then you can start using the mixedbreadai embedding model during vector table schema creation:
