VectorCode
CreatedDavidyz
Indexes code repositories to provide context to LLMs for improved code completion and understanding.
About
VectorCode is a code repository indexing tool designed to enhance the performance of Large Language Models (LLMs) when working with code. By indexing and providing relevant code repository information, VectorCode enables LLMs to generate more accurate and context-aware code completions and suggestions. It is particularly useful for closed-source projects, less known projects, and cutting-edge developments where LLMs may lack sufficient knowledge.
Key Features
- Indexes code repositories for LLM context
- Supports multiple embedding engines (e.g., SentenceTransformer)
- Includes a Neovim plugin for easy integration
- Offers chunking support with configurable chunk sizes
- Respects .gitignore files
- 207 GitHub stars
Use Cases
- Improving code completion accuracy with LLMs
- Providing context to LLMs for better understanding of codebases
- Assisting in the development of projects with limited LLM knowledge