I tried out Amazon Bedrock, and used Tonic Validate to do a head to head comparison of very simple RAG system's built using embedding and text models available in Amazon Bedrock. I compared Amazon Titan's embedding and text models to Cohere's embedding and text models in RAG systems that employ Amazon Bedrock Knowledge Bases as the vector db and retrieval components of the system.
The code for the comparison is in this jupyter notebook https://github.com/TonicAI/tonic_validate/blob/main/examples...
Let me know what you think, And your experiences building RAG with Amazon Bedrock!
