Skip to content

hellogreencow/chat

Repository files navigation

🤖 AI Chat Assistant

A super simple AI chat app that runs on your computer!

No complicated setup, no monthly fees, no data sent to the cloud. Just you, your computer, and powerful AI models running locally.

Main Interface The beautiful main chat interface - clean, simple, and powerful

🎯 What Does This Do?

Think of this as your personal ChatGPT that:

  • 🏠 Runs entirely on YOUR computer (private & secure)
  • 🧠 Analyzes your system and recommends the best AI models
  • 👁️ Can see and understand images you upload
  • 💻 Helps with coding, writing, questions - anything!
  • 📁 Handles huge files (up to 500GB!)
  • Works great on Apple Silicon Macs (M1, M2, M3, M4)

System Analysis Smart system detection - it knows your hardware and suggests the perfect models

🚀 Super Easy Setup (10 Minutes!)

Step 1: Get the Basic Tools (5 minutes)

  1. Install Node.js (if you don't have it):

    • Go to nodejs.org
    • Download the green "LTS" version
    • Install it (just click through the installer)
  2. Install Ollama (the AI brain):

    • Go to ollama.ai
    • Click "Download" for your system
    • Install it (drag to Applications on Mac, run installer on Windows)

Step 2: Get This App (2 minutes)

  1. Download this app:

    git clone https://github.com/hellogreencow/chat.git
    cd chat
    npm install

    Don't know how to use terminal? No problem!

    • Mac: Press Cmd + Space, type "terminal", press Enter
    • Windows: Press Win + R, type "cmd", press Enter
    • Then copy-paste the commands above one by one

Step 3: Start Everything (3 minutes)

  1. Start the AI server:

    ollama serve

    (Keep this window open - it's your AI brain running)

  2. In a NEW terminal window, start the app:

    npm run dev
  3. Open your browser and go to: http://127.0.0.1:5556

🎉 That's it! You now have your own AI assistant!

Model Settings Easy model management - the app connects to Ollama automatically

🤖 Getting Your First AI Model

The app will guide you, but here's the super simple version:

  1. Click the "System" button in the top right
  2. See what the app recommends for your computer
  3. Click "Install" on any model you like

Or manually install a model:

ollama pull llama3.2:8b

Confused about which model to choose? Here's the simple guide:

  • 🏃‍♂️ Fast & Light: llama3.2:3b (great for quick questions)
  • 🎯 Best Balance: llama3.2:8b (perfect for most people)
  • 👁️ Can See Images: llava:7b (upload photos and ask about them!)
  • 💻 Coding Helper: codellama:7b (writes and explains code)

Model Capabilities Different AI models for different tasks - the app explains everything

🎨 Cool Things You Can Do

📸 Upload Images and Ask Questions

  • Drag any photo into the chat
  • Ask "What's in this image?"
  • Get detailed descriptions, identify objects, read text in images

💻 Get Coding Help

  • Ask it to write code in any language
  • Debug your programs
  • Explain complex code concepts

📝 Writing Assistant

  • Write emails, essays, stories
  • Fix grammar and improve writing
  • Translate between languages

🧠 General Knowledge

  • Ask about history, science, current events
  • Get explanations for complex topics
  • Research assistance

📁 Analyze Documents

  • Upload text files, PDFs (up to 500GB!)
  • Summarize long documents
  • Extract key information

🔧 Troubleshooting (If Something Goes Wrong)

"The app won't start"

  • Make sure Node.js is installed: type node --version in terminal
  • Make sure you're in the right folder: cd chat

"No models available"

  • Make sure Ollama is running: ollama serve
  • Try installing a model: ollama pull llama3.2:3b

"Can't connect to Ollama"

  • Check if Ollama is running: look for it in your system tray
  • Restart Ollama if needed

"Images won't upload"

  • Make sure you have a vision model like llava:7b
  • The app will warn you if your current model can't see images

🍎 Special Notes for Mac Users

If you have an Apple Silicon Mac (M1, M2, M3, M4), you're in luck! This app:

  • 🔍 Automatically detects your chip
  • Runs models super efficiently
  • 🧠 Recommends models optimized for your Mac
  • 💾 Takes advantage of unified memory

The system analyzer will show your exact chip and recommend the best models for your setup.

🎯 What Makes This Special?

  • 🔒 100% Private: Your conversations never leave your computer
  • 💰 Free Forever: No subscriptions, no API keys, no limits
  • 🚀 Fast: Runs locally, no internet required for chatting
  • 🎨 Beautiful: Modern, clean interface that's joy to use
  • 🧠 Smart: Automatically optimizes for your hardware
  • 👁️ Multimodal: Text, images, documents - it handles everything

📚 Want to Learn More?

  • Ollama Models: ollama.ai/library
  • Ask Questions: Open an issue on GitHub
  • Contribute: Pull requests welcome!

Made with ❤️ for everyone who wants powerful AI without the complexity

No PhD in computer science required! 🎓

About

OpenSource Local Chat Interface

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published