U
unopenAIOpen Chat
Local AI • Your Computer • Your Model • Your Data

Run AI on your own machine

No mandatory proprietary AI API. unopenAI connects to Ollama running on localhost. Your messages stay on your hardware when in Local Mode.

Local AI Offline
http://localhost:11434 •

Setup Steps

  1. 01
    Install Ollama
    Download from ollama.com or brew install ollama
    brew install ollama
# or visit https://ollama.com/download
  2. 02
    Download an open model
    Qwen, Llama, DeepSeek, Mistral, etc.
    ollama pull qwen3
ollama pull llama3.1
ollama pull deepseek-r1
  3. 03
    Start Ollama
    Ollama runs an API on :11434
    ollama serve
# keep running in background
  4. 04
    Connect unopenAI
    Set OLLAMA_BASE_URL if not default
    OLLAMA_BASE_URL=http://localhost:11434
# in .env.local
  5. 05
    Chat
    Open /chat, select Auto or Local model
    /chat → Model: Auto → Local Connected

Platform Guides

macOS
brew install ollama, then ollama pull qwen3, ollama serve
Windows
Download installer from ollama.com, then PowerShell: ollama pull qwen3
Linux
curl -fsSL https://ollama.com/install.sh | sh, then ollama pull qwen3

Detected Local Models

No local models detected. Run ollama pull qwen3 to download one.
Privacy Indicator
LOCAL MODE: Your messages are processed by your local model. No mandatory cloud AI provider.