Getting Started¶
Installation¶
# pip
pip install salmalm
# pipx (recommended for CLI tools)
pipx install salmalm
# From source
git clone https://github.com/hyunjun6928-netizen/salmalm.git
cd salmalm
pip install -e .
First Launch¶
- Setup Wizard — Choose whether to set a master password
- Onboarding — Paste your API keys (at least one provider)
- Main UI — Start chatting at
http://localhost:18800
API Key Setup¶
You need at least one provider API key:
| Provider | Get API Key |
|---|---|
| Anthropic | console.anthropic.com |
| OpenAI | platform.openai.com |
| aistudio.google.com | |
| xAI | console.x.ai |
Local LLM (No API Key Needed)¶
# Start Ollama
ollama serve
ollama pull llama3.2
# In SalmAlm Settings → Local LLM
# Endpoint: http://localhost:11434/v1
CLI Options¶
salmalm # Default (port 18800)
salmalm --port 9000 # Custom port
salmalm --open # Auto-open browser
salmalm --shortcut # Create desktop shortcut
salmalm --update # Self-update from PyPI
Environment Variables¶
SALMALM_PORT=18800 # Web server port
SALMALM_BIND=127.0.0.1 # Bind address (0.0.0.0 for LAN)
SALMALM_HOME=~/SalmAlm # Data directory
Docker¶
git clone https://github.com/hyunjun6928-netizen/salmalm.git
cd salmalm
docker compose up -d
# → http://localhost:18800
Telegram Bot Setup¶
- Talk to @BotFather →
/newbot - Copy bot token
- In SalmAlm web UI → Settings → Telegram → Paste token + your chat ID
- Restart SalmAlm
Discord Bot Setup¶
- Discord Developer Portal → New Application
- Bot → Reset Token → Copy
- OAuth2 → bot + applications.commands → Generate URL → Add to server
- In SalmAlm web UI → Settings → Discord → Paste token
- Restart SalmAlm