The Privacy-First AI Pair Programmer for Visual Studio.
Bringing the power of local LLMs directly into your IDE with Ollama.
LocalPilot is a powerful Visual Studio extension that integrates local Large Language Models (LLMs) via Ollama. It provides a seamless, high-performance coding experience without the need for cloud-based subscriptions or data privacy concerns.
|
A dedicated side panel for complex reasoning, code generation, and deep-dive technical discussions.
|
Instant access to Refactor, Explain, or Document code directly from your right-click context menu.
|
|
Easily manage your Ollama connection and assign different models for chat and autocomplete tasks.
|
|
- 🔒 Absolute Privacy: Your source code stays on your machine. No telemetry, no cloud hooks, no data leakage. Perfect for enterprise and sensitive projects.
- ⚡ Zero Latency: No waiting for cloud API responses. Local inference provides near-instantaneous completions.
- 💰 One-time Setup, Zero Cost: No recurring subscriptions. Use the power of your own hardware to fuel your development.
- 🎨 Native Experience: Designed to feel like a built-in Visual Studio feature, supporting both Light and Dark themes natively.
You must have Ollama installed and running on your machine.
- Download: ollama.com
- Launch a Model: We recommend code-centric models like
llama3,codellama, orphi3.ollama run llama3
- Visit the Visual Studio Marketplace.
- Click Download, or search for "LocalPilot" within the Visual Studio Extension Manager:
- Extensions > Manage Extensions > Online
- Restart Visual Studio to complete the installation.
Navigate to Tools > Options > LocalPilot > Settings.
- Ollama Base URL: Usually
http://localhost:11434. Click "Test Connection" to verify. - Model Assignments: Assign preferred models for Chat and Inline Completions.
Tip
For optimal performance, use a lightweight model like phi3 or starcoder2:3b for Inline Completions, and a larger model like llama3:8b or deepseek-coder for the Chat Assistant.
Simply start typing in any supported file. LocalPilot will provide translucent "ghost-text" suggestions.
Tab: Accept the suggestion.Esc: Dismiss the suggestion.
The dedicated AI chat panel can be summoned at any time:
- Global Shortcut: Press
Alt + Lto toggle the chat window. - LocalPilot Menu: Access Chat and Settings directly from the top-level LocalPilot menu or the Tools > LocalPilot menu in Visual Studio.
Right-click on any code selection or use the LocalPilot menu to access:
- Explain Code: Breakdown complex logic.
- Generate Docs: Auto-generate XML/docstring comments.
- Refactor: Suggest improvements for readability and performance.
We welcome community contributions! Whether it's bugs, features, or documentation, your help is appreciated.
- Check Issues: See the Existing Issues to avoid duplicates.
- Clear Reports: For bugs, include your VS version, Ollama model, and reproduction steps.
- Pull Requests: Create a branch from
main, ensure the project builds, and submit your PR with a clear description.
Since LocalPilot runs Large Language Models (LLMs) entirely on your local machine via Ollama, your hardware performance directly impacts the speed and responsiveness of AI suggestions.
- CPU: Recent Multi-core processor (Intel i5/AMD Ryzen 5 or equivalent).
- RAM: 8GB (16GB+ strongly recommended for a smooth experience).
- GPU: 4GB VRAM (Dedicated NVIDIA or Apple Silicon GPU preferred for faster inference).
- Storage: 5GB+ for model storage (SSD/NVMe highly recommended).
- RAM: 32GB+ for handling larger models (13B+) alongside Visual Studio.
- GPU: NVIDIA RTX 3060/4060 or higher with 12GB+ VRAM.
- NVIDIA CUDA: Ensure latest drivers are installed for GPU acceleration.
Important
LocalPilot is designed for efficiency, but because it performs all AI processing locally, it requires capable hardware. If suggestions feel slow, consider using a smaller, quantized model (e.g., phi3:mini or starcoder2:3b) in the settings.
"A major leap in visual fidelity and user experience refinement."
- ✨ High-Fidelity Syntax Highlighting: New theme-aware code renderer with a custom regex engine for premium visualization of methods, types, and strings.
- 🎨 Theme-Aware Palette: Deep integration with VS Dark and Light themes—colors automatically adapt to your environment for maximum readability.
- ⌨️ Global Access Shortcut: Added
Alt + Las the universal command to instantly summon the AI Chat panel. - 📂 Dedicated LocalPilot Menu: A new top-level LocalPilot menu is now available (also accessible via Tools > LocalPilot) for rapid access to chat and settings.
- 🖼️ Minimalist Branding: Refined the chat interface by removing redundant text labels, focusing purely on a clean, icon-centric AI persona.
- 🧠 Intelligent Status Tracking: Resolved the "sticky thinking" bug; the AI now correctly transitions from "thinking" to "worked for X.Xs" without clutter.
- 🛡️ Robust Stream Rendering: Hardened the UI update loop against rapid-fire AI streaming, preventing flickering and ensuring smooth output.
"The production-ready overhaul focusing on reliability and UX."
- ✅ Selection Deadlock Fix: Fundamental refactor of selection capture to prevent UI thread hangs during Quick Actions.
- ✅ Live Markdown Engine: New incremental renderer provides a premium "typing" experience with real-time headers and list formatting.
- ✅ Hardened Container Logic: Resolved critical
NullReferenceExceptionandObjectDisposedExceptionerrors during intensive AI streaming. - ✅ Thread-Safe Diagnostics: Refactored the internal logger to be non-blocking and safe for background AI operations.
- ✅ Pro Marketplace Branding: Formally aligned all project identifiers with the official
FutureStackSolutionnamespace.
- Inline Ghost-Text: Real-time code completions via Ollama.
- Interactive Chat Panel: Full technical discussion window.
- Context Menu Actions: Explain, Document, and Refactor support.
- Native VS Support: Full theme awareness for Light and Dark modes.