Using Perplexity Pro as an Integration with AI Model

Yes, you can use Perplexity as an integration in Clay. We have a native Perplexity AI integration that allows you to search the web for information with AI or use a non-online model. To access it, click Add Enrichment and search for "Perplexity AI". Each enriched cell costs 1 credit. You can select different Perplexity models in the Model dropdown menu based on your needs. The integration works on all Clay plans (Starter, Explorer, Pro, Enterprise) and uses Clay's shared API.

Sources: Perplexity integration overview | Documentation | Clay University Perplexity AI x Clay integration | Clay.com Online AI search with Perplexity AI | Clay.com

Yes, you can use your own Perplexity Pro account to avoid using Clay credits. When you bring your own API key for any integration, you won't be charged Clay credits. To add your Perplexity API key: 1. Go to Settings > Connections 2. Click Add Connection and search for Perplexity 3. Configure with your API key Alternatively, when setting up a Perplexity enrichment, select Add Account to connect your own API key instead of using Clay's shared account. This feature is available on paid Clay plans (Starter, Explorer, Pro, Enterprise).

Sources: Do I get charged credits if I bring my own API key? - FAQ | Clay.com Managing integration accounts | Documentation | Clay University Guide: Ways to save Clay credits | Documentation | Clay University

The Perplexity integration doesn't have built-in JSON response customization like Claygent. However, you can achieve customizable responses by using the System Prompt field to guide the model's output format. In the Perplexity integration setup, use the System Prompt field to specify that you want JSON-formatted responses with your desired structure. For example, you could prompt it to return data in specific JSON fields. Alternatively, if you need more control over JSON responses, consider using the HTTP API integration to directly call Perplexity's API, which gives you full control over request/response formatting and allows you to define specific field paths to return from the JSON response. Could you copy and paste the specific JSON format you're trying to achieve? This will help troubleshoot the best approach for your use case.

Sources: Perplexity integration overview | Documentation | Clay University Perplexity integration overview | Documentation | Clay University HTTP API integration overview | Documentation | Clay University

The Perplexity integration doesn't have built-in JSON output formatting like Claygent. However, you can use the System Prompt field to specify your desired JSON structure. In the System Prompt field, instruct the model to return responses in your exact JSON format: LBKSReturn the response in this exact JSON format: { "company_size_rule": "<one-sentence rule statement> ", "prospects": [ { "full_name": "<Full Name> ", "title": "<Job Title> ", "linkedin_url": "<URL> ", "reasoning": "<≤ 25-word justification> " } ] }LBKSAlternatively, use the HTTP API integration to directly call Perplexity's API for full control over JSON request/response formatting. Since you mentioned JSON formatting, could you copy and paste your current setup so I can troubleshoot any specific issues?

Sources: Perplexity integration overview | Documentation | Clay University Perplexity integration overview | Documentation | Clay University HTTP API with JWT authentication | Documentation | Clay University

Hey Jacob! It looks like your question was answered, but happy to help if anything else comes up! For now, I'll close out this thread, but feel free to respond or start a new one!