Skip to content
Nappai Documentation

Perplexity

Generate creative text using powerful AI language models. The Perplexity component in Nappai lets you create text using advanced AI. You can customize the output to fit your needs by adjusting various settings.

Relationship with Perplexity AI

This component uses the Perplexity AI language models to generate text. It connects to the Perplexity API to access these powerful models and allows you to control how the text is generated. You’ll need a Perplexity API key to use this component.

Inputs

  • Model Name: Choose the AI model you want to use. Different models offer different strengths and styles of writing. The default option is a good starting point.
  • Max Output Tokens: Set the maximum length of the generated text. A token is roughly equivalent to a word. Increase this number for longer text.
  • Perplexity API Key: This is a secret key that allows Nappai to access the Perplexity AI service. Keep this key confidential. You’ll get this key from your Perplexity AI account.
  • Temperature: Controls how creative the AI is. A lower temperature (e.g., 0.2) makes the text more predictable and focused. A higher temperature (e.g., 1.0) makes it more creative and unpredictable. The default is 0.75, which is a good balance.
  • Top P (Advanced): A more advanced setting that further controls the randomness of the output. Leave this at the default unless you have a specific reason to change it.
  • N (Advanced): If you’re using a chat-based model, this determines how many different responses the AI will generate for the same prompt.
  • Top K (Advanced): Another advanced setting that influences the selection of words during text generation. Leave this at the default unless you’re familiar with this technique.

Outputs

The component doesn’t have explicit outputs in the visual dashboard. Instead, the generated text is directly used by other components in your Nappai workflow. For example, the generated text might be used as input for a summarizer or a translator.

Usage Example

Let’s say you want to generate a short product description for a new type of coffee. You would:

  1. Select a suitable “Model Name” (perhaps llama-3.1-sonar-small-128k-online).
  2. Set “Max Output Tokens” to, say, 50.
  3. Enter your Perplexity API Key.
  4. (Optional) Adjust “Temperature” to control the creativity of the description.
  5. Provide a prompt like “Write a short, catchy product description for a new dark roast coffee with notes of chocolate and hazelnut.”
  6. The generated text will then be available for use by other components in your workflow.

Templates

[List of templates where the component is used will be added here once available.]

  • Summarizer: Use the Perplexity component to generate text, then use the Summarizer to shorten it.
  • Translator: Generate text in one language using Perplexity, then translate it using the Translator component.
  • Explain SQL results: Use Perplexity to generate a human-readable explanation of complex SQL query results.

Tips and Best Practices

  • Start with the default settings and adjust them gradually to find what works best for your needs.
  • Experiment with different model names to see which one produces the best results for your specific task.
  • Always keep your Perplexity API key safe and secure.

Security Considerations

  • Never share your Perplexity API key with anyone. Treat it like a password.
  • Regularly review the security settings of your Perplexity AI account.
  • Be mindful of the data you use as input to the Perplexity component, ensuring it doesn’t contain sensitive information.