Quickstart: configuration

Load and manage your Attune AI configuration with a single function call.

from attune.config.loader import load_unified_config

config = load_unified_config()
print(f"Model: {config.model_id}")
print(f"Provider: {config.provider}")

Set up your configuration

1. Create a configuration file in one of these locations:

   • ./attune.config.json (project directory)
   • ~/.attune/config.json (user directory)
   • ~/.config/attune/config.json (XDG config directory)

2. Add basic settings to your JSON file:

   {
     "provider": "openai",
     "model": "gpt-4",
     "temperature": 0.7,
     "max_tokens": 2000
   }
   

3. Load and verify your configuration:

   from attune.config.loader import load_unified_config
   
   config = load_unified_config()
   print(f"Using {config.provider} with model {config.model_id}")
   

Expected output:

Using openai with model gpt-4

You can also override any setting using environment variables with the ATTUNE_ prefix, like ATTUNE_MODEL=gpt-3.5-turbo.

Next steps

Configure Redis integration for state management by adding a redis section to your config file.