KDEGroup/CodeRAG
Source code for EMNLP'25 paper "CodeRAG: Finding Relevant and Necessary Knowledge for Retrieval-Augmented Repository-Level Code Completion".
This project helps evaluate advanced code completion systems that provide suggestions based on a whole codebase, not just individual files. It takes a benchmark dataset of code completion scenarios and outputs performance metrics. This tool is designed for researchers and developers working on improving and testing large-scale code completion models.
Use this if you are a researcher or developer working on retrieval-augmented code completion and need to benchmark your models against established datasets.
Not ideal if you are an end-user developer looking for an IDE plugin or a simple code completion tool.
Stars
18
Forks
—
Language
Python
License
MIT
Category
Last pushed
Jan 18, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/rag/KDEGroup/CodeRAG"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
Kain-90/RAG-Play
An interactive visualization tool for understanding Retrieval-Augmented Generation (RAG) pipelines.
rryam/LumoKit
Swift package for on-device Retrieval-Augmented Generation (RAG)
harvard-lil/warc-gpt
WARC + AI - Experimental Retrieval Augmented Generation Pipeline for Web Archive Collections.
CoIR-team/coir
(ACL 2025 Main) A Comprehensive Benchmark for Code Information Retrieval.
constacts/ragtacts
RAG(Retrieval-Augmented Generation) for Evolving Data