Benchmarking RAG on tables

Key links

LangChain public benchmark evaluation notebooks:

• Long context LLMs here
• Chunk size tuning here
• Multi vector with ensemble here

Motivation

Retrieval augmented generation (RAG) is one of the most important concepts in LLM app development. Documents of many types can be passed into the context window

app benchmark benchmarking concepts context context window development documents ensemble evaluation llm llms notebooks public rag tables types vector