Skip to content

IfAI (若爱)AI-Native Code Editor

The autonomous programming companion for the AI era

IfAI Logo

Why IfAI?

In the AI era, code editors should be more than containers for text—they should be the physical body of AI intelligence.

IfAI is built from the ground up as an AI-native development environment where reasoning capabilities are integrated into the kernel, not added as plugins.

Key Features

🧠 AI-Native Core

  • Deep integration of AI capabilities into the editor kernel
  • Conversation-aware code context and file operations
  • Intelligent routing between local and cloud models

⚡ Extreme Performance

  • Rust-powered backend for native performance
  • 120 FPS smooth rendering
  • Sub-15ms UI response latency

🔒 Privacy First

  • Local LLM support via Ollama
  • Sensitive code never leaves your machine
  • Full offline capability

🤖 Autonomous Agents

  • Shell-level control for real task execution
  • Automatic environment configuration
  • Self-correction and error recovery

🎯 Symbol-Aware RAG

  • Tree-sitter based code parsing
  • Cross-file symbol relationship analysis
  • Accurate code structure understanding

👁️ Multimodal Vision

  • Drag-and-drop image upload
  • Screenshot paste support (Cmd+V)
  • AI understands UI designs and diagrams

Quick Start

Installation

bash
# Clone the repository
git clone https://github.com/peterfei/ifai.git
cd ifai

# Install dependencies
npm install

# Start development
npm run tauri dev

Your First AI Conversation

  1. Open IfAI
  2. Press Cmd+K to open the AI chat panel
  3. Type your question or request
  4. Watch AI understand your codebase context

What's New

v0.3.0 - Multimodal Intelligence

  • Vision LLM Support: Drag and drop images for AI analysis
  • Local/Remote Hybrid Routing: Smart model selection
  • Code Navigation: F12 Go to Definition, Find References
  • AI Refactoring: Automatic code quality improvements
  • New User Tour: Interactive onboarding experience

View full changelog →

Community

License

MIT License - see LICENSE for details.

Released under the MIT License.