Boost coding with GitHub Copilot Free for VS Code 1.136—tab complete, chat, agent host, free sign-in.
Proxy the GitHub Copilot API to support custom integrations through a reliable, rewritten interface.
🧠 Generate dynamic narratives with LMD, a framework for episodic memory that evolves, feels, and tells stories through creative language grounding.
Enable GitHub Copilot access through a free C++ proxy that routes requests to alternative endpoints and blocks telemetry
Run a local coding agent with llama.cpp for private, containerized development that supports CPU and NVIDIA GPU setups without external APIs.
🚀 Streamline access to major AI models like ChatGPT and Claude with a unified API, enhancing reliability and reducing costs for developers in China.
Documentation for the twinny Visual Studio Code extension
Compress context data to optimize memory and performance in C++ large language model applications within the llm-cpp toolkit.
LMDeploy is a toolkit for compressing, deploying, and serving LLMs.
Python Co-Pilot enhances Python developers' coding experience by utilizing LLama's advanced language understanding and Mistral's intelligent code generation, enabling more efficient and high-quality code writing.
Comparison of LoRA and Knowledge Editing for Improving Neural Decompilation
Local CodeLlama workbench for debugging, testing, refactoring, explanation and security review.
A coding assistant that never leaves home. Explains and completes code using a code-specialised LLM running locally via Ollama (DeepSeek-Coder, CodeLlama). Native fill-in-the-middle, language detection, SQLite caching, Streamlit UI. No API keys, no network calls.
RAG pipeline using CodeLlama for code generation and explanation
Leveraged a Large Language Model query system to convert natural language requests into executable MongoDB queries, reducing dependency on manual query writing and enabling teams to access business data faster, improve operational efficiency, and make informed decisions.
Interactive web-dev learning platform — PHP / MySQL, Monaco editor, CodeLlama assistant. Live demo.
Self-hosted semantic code search (RAG) with orchestration, audit logging, and prompt-injection guarding — evaluated at 87.5% top-3 accuracy.
"Multi-agent LLM system achieving 99.4% success rate on automated security vulnerability repair across 8 CWE categories"
To associate your repository with the codellama topic, visit your repo's landing page and select "manage topics."