Dataset · onyx-dot-app
Benchmark for evaluating corporate RAG
onyx-dot-app/EnterpriseRAG-Bench
Over 500,000 documents and 500 English questions for testing RAG systems on business data.
What's inside
An English text benchmark including over 500,000 realistic corporate documents and 500 questions. Unlike common open-web datasets, this data mimics a closed company working environment.
What you can do with it
- Evaluating search and response generation quality (RAG) in corporate systems
- Comparing retrievers and rerankers on complex business documents
- Tasks
- Question answering, Text retrieval
- Languages
- English
- Size
- up to 1K rows
- Downloads this month
- 2.1K
- Likes
- 27
How to load it
from datasets import load_dataset
ds = load_dataset("onyx-dot-app/EnterpriseRAG-Bench")