rungalileo/hallucination-index

Initiative to evaluate and rank the most popular LLMs across common task types based on their propensity to hallucinate.

31
/ 100
Emerging

Evaluates 22 models across three RAG scenarios (short/medium/long context) using ChainPoll—a multi-polling chain-of-thought technique—to quantify hallucinations and contextual adherence. Tests both open and closed-source models against variable context lengths (5k-100k tokens) and prompting strategies like Chain-of-Note. Includes custom LLM-based evaluation for factual accuracy and position-bias analysis across 10,000 domain-specific documents.

116 stars. No commits in the last 6 months.

No License Stale 6m No Package No Dependents
Maintenance 2 / 25
Adoption 10 / 25
Maturity 8 / 25
Community 11 / 25

How are scores calculated?

Stars

116

Forks

9

Language

License

Last pushed

Jul 28, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/rag/rungalileo/hallucination-index"

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