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.
The beautiful main chat interface - clean, simple, and powerful
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)
Smart system detection - it knows your hardware and suggests the perfect models
-
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)
-
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)
-
Download this app:
git clone https://github.com/hellogreencow/chat.git cd chat npm installDon'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
- Mac: Press
-
Start the AI server:
ollama serve
(Keep this window open - it's your AI brain running)
-
In a NEW terminal window, start the app:
npm run dev
-
Open your browser and go to:
http://127.0.0.1:5556
🎉 That's it! You now have your own AI assistant!
Easy model management - the app connects to Ollama automatically
The app will guide you, but here's the super simple version:
- Click the "System" button in the top right
- See what the app recommends for your computer
- Click "Install" on any model you like
Or manually install a model:
ollama pull llama3.2:8bConfused 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)
Different AI models for different tasks - the app explains everything
- Drag any photo into the chat
- Ask "What's in this image?"
- Get detailed descriptions, identify objects, read text in images
- Ask it to write code in any language
- Debug your programs
- Explain complex code concepts
- Write emails, essays, stories
- Fix grammar and improve writing
- Translate between languages
- Ask about history, science, current events
- Get explanations for complex topics
- Research assistance
- Upload text files, PDFs (up to 500GB!)
- Summarize long documents
- Extract key information
"The app won't start"
- Make sure Node.js is installed: type
node --versionin 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
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.
- 🔒 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
- 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! 🎓