---
title: Quran Colbert Retrieval Demo
emoji: 🌍
colorFrom: gray
colorTo: blue
sdk: gradio
sdk_version: 4.38.1
app_file: app.py
pinned: false
license: other
---

# Quran Colbert Retrieval Demo 
This demo is based on the simple text of the Quran and uses the Arabic ColBERT model trained on 100K Arabic (translated) triplets 
from the curated dataset (akhooli/arabic-triplets-1m-curated-sims-len). It is not intended as a replacement of scholar work, just to demo the 
performance of Arabic colBERT models on edge cases. 

You may pick an example or type your own query (this is not a chat app, it just retrives the most relevant parts of suras). Scroll up/down the output 
section (3 results per query, first is usually most relevant). Try to re-word a query if initial results are not relevant. 

# Related posts:  
https://www.linkedin.com/posts/akhooli_arabic-1-million-curated-triplets-dataset-activity-7222951839774699521-PZcw  
https://www.linkedin.com/posts/akhooli_heres-a-basic-demo-of-the-arabic-colbert-activity-7220100000721051649-NBfr  
https://www.linkedin.com/posts/akhooli_testing-the-first-arabic-colbert-semantic-activity-7219233042945495040-qg3I  
https://www.linkedin.com/posts/akhooli_this-is-probably-the-first-arabic-colbert-activity-7217969205197848576-l8Cy