aikho/retrivex

Explainability toolkit for retrieval models. Explain prediction of vector search models (embeddings similarity models, siamese encoders, bi-encoders, dense retrieval models). Debug your vector search models for RAG or agentic AI system.

23
/ 100
Experimental

This toolkit helps you understand why your AI system retrieves specific information when it’s asked a question or given a prompt. You input a query and a retrieved document, and it shows you which parts of each were most responsible for the match. It's designed for AI system developers or engineers building applications like intelligent chatbots or advanced search engines.

No commits in the last 6 months.

Use this if you need to debug, build trust in, or improve the accuracy of your information retrieval system by understanding why certain text pairs are considered similar.

Not ideal if you're trying to explain traditional classification or regression model outputs, as it's specifically tailored for vector search and similarity models.

AI-debugging information-retrieval RAG-systems vector-search explainable-AI
Stale 6m No Package No Dependents
Maintenance 2 / 25
Adoption 6 / 25
Maturity 15 / 25
Community 0 / 25

How are scores calculated?

Stars

15

Forks

Language

Jupyter Notebook

License

LGPL-2.1

Last pushed

Oct 05, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/embeddings/aikho/retrivex"

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