QUICK START

Quick Start

Install Chalie

curl -fsSL https://chalie.ai/install | bash

The installer:

  1. Checks prerequisites (Python 3.9+, Docker optional)
  2. Downloads the latest release and builds in place (~2 min)
  3. Installs the chalie CLI and opens Chalie at http://localhost:8081

No root access required. Everything lives in ~/.chalie/.

After Install

chalie                 # Start Chalie → http://localhost:8081
chalie --port=9000     # Start on a custom port
chalie stop            # Stop the process
chalie restart         # Restart
chalie update          # Update to the latest release
chalie status          # Check if running
chalie logs            # Follow the log

Onboarding

Open http://localhost:8081/on-boarding/ and:

  1. Create an account — set a password
  2. Configure an LLM provider — choose from the options below
  3. Begin — you’ll be redirected to the chat interface

LLM Providers

Recommended setup:

  • Local / private — Install Ollama, run ollama pull qwen:8b, point Chalie at http://localhost:11434. Free, no data leaves your machine.
  • Easiest cloud — Use OpenAI or Anthropic. Paste an API key during onboarding and you’re done.

Ollama (local)

# Install from https://ollama.ai, then:
ollama pull qwen:8b

In onboarding, select Ollama and set the endpoint to http://localhost:11434.

OpenAI

  1. Get an API key from platform.openai.com
  2. In onboarding, select OpenAI and paste your key

Anthropic

  1. Get an API key from console.anthropic.com
  2. In onboarding, select Anthropic and paste your key

Google Gemini

  1. Get an API key from ai.google.dev
  2. In onboarding, select Gemini and paste your key

Configuration

All configuration (LLM providers, API keys, settings) is done via the web UI after first run. The only runtime option is the port:

chalie --port=9000     # Start on a custom port (default: 8081)

Voice features auto-detect native dependencies — no Docker needed. When voice deps are installed (via the installer or pip install -r backend/requirements-voice.txt), voice appears automatically. When they’re not, voice is silently hidden. Use --disable-voice during install to skip voice dependencies entirely.

Updating

chalie update

Re-runs the installer with CHALIE_UPDATE=1: stops the running process, downloads the latest source, reinstalls dependencies. Your database and memory in ~/.chalie/data/ are never touched.

Uninstalling

chalie stop
rm -rf ~/.chalie ~/.local/bin/chalie

Remove the export PATH="$HOME/.local/bin:$PATH" line from ~/.bashrc or ~/.zshrc if it was added by the installer.

For Hackers & Contributors

Want to run from source, patch internals, or contribute?

Prerequisites: Python 3.9+, git

Steps:

1. Clone the repo

git clone https://github.com/chalie-ai/chalie.git
cd chalie

2. Create a virtual environment and install dependencies

python3 -m venv .venv
source .venv/bin/activate
pip install -r backend/requirements.txt

3. Run Chalie

python backend/run.py
# Opens at http://localhost:8081

4. Run tests

cd backend && pytest

Port override (optional):

python backend/run.py --port=9000

All other configuration (LLM providers, API keys) is done via the web UI after first run. Voice auto-detects native dependencies (no Docker needed).

See CONTRIBUTING.md for contribution guidelines.