naver/bergen

Benchmarking library for RAG

55
/ 100
Established

Provides modular evaluation of RAG pipelines across 20+ retrievers, 4 rerankers, and 20+ LLMs with YAML-based configuration and caching. Includes LLM-as-judge evaluation, multilingual support, and established baselines on KILT and QA datasets (ASQA, NQ, TriviaQA, HotPotQA). Integrates with VLLM for generation and supports training extensions via LoRA.

261 stars.

No Package No Dependents
Maintenance 13 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 16 / 25

How are scores calculated?

Stars

261

Forks

31

Language

Jupyter Notebook

License

Last pushed

Mar 11, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/rag/naver/bergen"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.