Available Embedding Models

A number of embedding models are available to be used with the Private AI Services Container, including some that are shipped with the container. There are also pre-built models available for download as well as models that are known to work with the container but must be built using Oracle Machine Learning for Python (OML4Py) Client 2.1.

The following models are included with the container and are available without any additional download or build steps:

  • all-mpnet-base-v2
  • all-MiniLM-L12-v2
  • multilingual-e5-base
  • multilingual-e5-large
  • clip-vit-base-patch32-txt
  • clip-vit-base-patch32-img

The following ONNX Pipeline Model is pre-built and available for download. You can find the link for downloading the model at Oracle Machine Learning for Python User’s Guide:

  • multilingual-e5-small

The following lists of models are ordered by popularity based on download frequency. They are known to work with the container but must be built using OML4Py Client 2.1. Once you have built your desired ONNX Pipeline model, you must copy that ONNX file into a directory on the host machine in which the container will run. These models can be found for download at https://huggingface.co/models.

English Vector Embedding Models:

  • sentence-transformers/all-MiniLM-L6-v2
  • sentence-transformers/all-mpnet-base-v2
  • prajjwal1/bert-tiny
  • BAAI/bge-base-en-v1.5
  • BAAI/bge-small-en-v1.5
  • colbert-ir/colbertv2.0
  • sentence-transformers/all-MiniLM-L12-v2
  • BAAI/bge-large-en-v1.5
  • nomic-ai/nomic-embed-text-v1.5
  • mixedbread-ai/mxbai-embed-large-v1
  • sentence-transformers/multi-qa-MiniLM-L6-cos-v1
  • thenlper/gte-large
  • intfloat/e5-base-v2
  • intfloat/e5-large-v2
  • WhereIsAI/UAE-Large-V1
  • Snowflake/snowflake-arctic-embed-xs
  • thenlper/gte-base
  • Snowflake/snowflake-arctic-embed-m
  • thenlper/gte-small
  • intfloat/e5-small-v2
  • Snowflake/snowflake-arctic-embed-s
  • jinaai/jina-embeddings-v2-small-en
  • jinaai/jina-embeddings-v2-base-en
  • TaylorAI/bge-micro-v2
  • Snowflake/snowflake-arctic-embed-m-v1.5
  • TaylorAI/gte-tiny
  • Snowflake/snowflake-arctic-embed-l

Multilingual Text Embedding Models:

  • sentence-transformers/paraphrase-multilingual-mpnet-base-v2
  • intfloat/multilingual-e5-small
  • sentence-transformers/distiluse-base-multilingual-cased-v2
  • intfloat/multilingual-e5-base
  • sentence-transformers/multi-qa-MiniLM-L6-cos-v1
  • sentence-transformers/stsb-xlm-r-multilingual
  • ibm-granite/granite-embedding-278m-multilingual
  • ibm-granite/granite-embedding-107m-multilingual

Vision Embedding Models:

  • openai/clip-vit-base-patch32
  • google/vit-base-patch16-224
  • facebook/deit-tiny-patch16-224
  • microsoft/resnet-50
  • microsoft/resnet-18
  • WinKawaks/vit-small-patch16-224
  • WinKawaks/vit-tiny-patch16-224

Cross-Encoder Embedding Models:

These models can be used for reranking.

  • cross-encoder/ms-marco-MiniLM-L6-v2
  • cross-encoder/ms-marco-MiniLM-L12-v2
  • BAAI/bge-reranker-base

Parent topic: Considerations when Configuring the Private AI Services Container