Perplexity AI

The BuildShip allows you to leverage the power of Perplexity AI's language models directly within your BuildShip workflows. Perplexity AI is designed to provide high-quality, coherent, and contextually relevant responses by leveraging cutting-edge Natural Language Processing (NLP) techniques.

BuildShip offers a pre-built Perplexity AI Node that simplifies the integration of Perplexity AI's language models into your workflows.

Prerequisites ✅

Perplexity API Key

To use the Perplexity AI nodes, you need a Perplexity API key. Follow the steps below to get your API key:

• Sign up for a Perplexity AI account at Perplexity AI (opens in a new tab), if you don't already have one.
• Navigate to your account settings. Under the API Section (opens in a new tab), register your credit card to get started. This step will not charge your credit card. Rather, it stores payment information for later API usage.
• Generate an API key. This key will be required for authentication in the Perplexity AI nodes.

Node: Perplexity AI

The Perplexity AI Node is a pre-built node that simplifies the integration of Perplexity AI's language models into your BuildShip workflows. The Perplexity AI Node can be found in the Node Library, under the Perplexity AI integration.

The node inputs allows you to specify various configuration parameters to customize the interaction with the Perplexity AI Chat API. The Perplexity AI Node accepts the following input parameters:

API Key

Your Perplexity AI API Key, which you can generate through the Perplexity Account Settings Page. Refer to the Prerequisites section for more information on obtaining your API Key.

System Prompt

The system's prompt message, which sets the context or instructions for the AI model.

Sample System Prompt:

You are a helpful kitchen assistant. You'll be asked to provide recipes for various dishes.

User Input

The user's prompt question or input text. Sample Input:

Sample User Input:

What is the recipe for a classic Margherita pizza? I have pizza dough, tomato sauce, mozzarella cheese, garlic cloves, and fresh basil.

Models

The Perplexity AI model to be used for generating the response. You can select from the following options:

• Mistral 7B Instruct: A large-scale language model trained on a diverse range of text sources, designed for generating instructional text.

• Mixtral 8X7B Instruct: A larger-scale language model trained on a diverse range of text sources, designed for generating instructional text.

• Sonar Small Chat: A smaller-scale language model trained on conversational text, designed for generating chatbot responses.

• Sonar Medium Chat: A medium-scale language model trained on conversational text, designed for generating chatbot responses.

• Sonar Small Online: A smaller-scale language model trained on online text, designed for generating online content.

• Sonar Medium Online: A medium-scale language model trained on online text, designed for generating online content.

• CodeLlama 70B Instruct: A large-scale language model trained on instructional text, designed for generating instructional content.

Output

The Perplexity AI Node directly returns the response generated by the Perplexity AI model based on the user input and system prompt. The response can be used in subsequent nodes within your workflow for further processing or analysis.