Awesome open coding-copilot and friends

This is a curated list of resources related to coding-copilot and related tools.

Table of Contents

• Awesome open coding-copilot and friends
• Table of Contents
• Plugins
• Models
  • Fine-tuning StarCoder
• Benchmarks/Datasets
  • Natural Language to Code
  • Code to Natural Language
  • Code Completion
  • Code Search
• Papers
• Performance Comparisons
• Related Open Source Projects
• Ethics and Challenges
• Contributing Guidelines

Plugins

These plugins integrate AI-assisted coding capabilities into various development environments:

• copilot-clone - An open-source alternative to GitHub Copilot
• fauxpilot - An open-source alternative to GitHub Copilot server
• twinny - The most no-nonsense, locally or API-hosted AI code completion plugin for Visual Studio Code
• claude-dev - Autonomous software engineer right in your IDE, capable of creating/editing files, executing commands, and more with your permission every step of the way.

Models

These are some of the prominent models used for code generation and understanding:

• CodeT5 - Open-source model for code understanding and generation
• CodeGen - Large language model for program synthesis
• StarCoder - Large language model trained on source code
• CodeBERT - Pre-trained model for programming language
• GPT-Neo - Open-source alternative to GPT-3
• CodeParrot - Large language model trained on code

Fine-tuning StarCoder

• StarCoder Fine-tuning Guide

Benchmarks/Datasets

Resources for evaluating and training code models:

• CodeSearchNet - Dataset and benchmark for code search
• Stack Exchange Dataset - Instruction dataset based on Stack Exchange
• The Pile - Large-scale dataset including programming language data
• APPS - Benchmark for code generation
• HumanEval - Benchmark for evaluating language models on coding tasks
• CodeXGLUE - Benchmark dataset for code intelligence

Natural Language to Code

• CoNaLa - Dataset for mapping natural language to code
• NL2Code - A dataset for natural language to code generation

Code to Natural Language

• CodeNN - Dataset for code summarization
• FunCom - Dataset for method name generation

Code Completion

• PY150 - Dataset for Python code completion
• CodeCompletionBenchmark - Google's benchmark for code completion

Code Search

• CodeSearchNet Challenge - Multiple datasets for code search tasks
• AdvTest - Advanced code-to-code search dataset

Papers

Seminal and recent research papers in the field:

• Evaluating Large Language Models Trained on Code - OpenAI's paper on Codex
• PyMT5: Multi-mode Translation of Natural Language and Python Code with Transformers
• InCoder: A Generative Model for Code Infilling and Synthesis
• Competition-Level Code Generation with AlphaCode
• CodeXGLUE: A Benchmark Dataset and Open Challenge for Code Intelligence
• CodeSearchNet Challenge: Evaluating the State of Semantic Code Search
• CodeBERT: A Pre-Trained Model for Programming and Natural Languages
• CodeT5: Transformer-based Models for Code Understanding
• StarCoder: A Large Language Model for Program Synthesis
• CodeParrot: A Large Language Model for Code Generation
• GPT-Neo: Large-Scale Language Models for Code Generation
• DeepCoder: Learning to Write Programs
• Code2Vec: Learning Distributed Representations of Code

Performance Comparisons

Studies and articles comparing the performance of different AI coding assistants:

• Comparing Copilot, ChatGPT, and Human Developers - Research study on code generation performance
• Evaluating LLM Performance on Programming Tasks - Comprehensive evaluation of various models

Related Open Source Projects

Open-source projects that complement or enhance AI-assisted coding:

• LSP - Language Server Protocol for editor-agnostic tooling
• Tree-sitter - Parser generator tool and incremental parsing library
• SonarQube - Static code analysis tool that can be enhanced with AI capabilities

Ethics and Challenges

Discussions and resources on the ethical considerations and challenges in AI-assisted coding:

• The Ethics of AI-Assisted Coding - ACM article on ethical considerations
• Challenges in AI-Assisted Coding - Research paper on challenges and future directions
• Copyright and AI-Generated Code - Legal perspective on AI-generated code

Contributing Guidelines

We welcome contributions to this awesome list! Here's how you can contribute:

1. Fork the repository
2. Create a new branch for your additions
3. Add your links and descriptions, following the existing format
4. Ensure your additions are in alphabetical order within their respective sections
5. Create a pull request with a clear description of your changes

Please make sure any resources you add are:

• Relevant to AI-assisted coding
• Of high quality and useful to the community
• Not duplicates of existing entries

Contributions to this awesome list are welcome! Please submit a pull request or open an issue to suggest additions or changes.