Listen to NavamAI Podcast

Privacy Controls

You decide which model and provider you trust, or even choose to run an LLM locally on your laptop. You are in control of how private your data and preferences remain. NavamAI supports state of the art models from Anthropic, OpenAI, Google, and Meta. You can choose a hosted provider or Ollama as a local model provider on your laptop. Switch between models and providers using a simple command like navamai config ask model llama to switch from the current model.

You can also load custom model config sets mapped to each command. Configure these in navamai.yml file. Here is an example of constraining how navamai ask and navamai intents commands behave differently using local and hosted model providers.

ask:
  provider: ollama
  model: mistral
  max-tokens: 300
  save: false
  system: Be crisp in your response. Only respond to the prompt 
    using valid markdown syntax. Do not explain your response.
  temperature: 0.3

intents:
  provider: claude
  model: sonnet
  max-tokens: 1000
  save: true
  folder: Embeds
  system: Only respond to the prompt using valid markdown syntax.
    When responding with markdown headings start at level 2. 
    Do not explain your response.
  temperature: 0.0