Playbooks perplexity-ai
perplexity-ai perplexity-ai

Conduct Accurate Deep Research Using Perplexity AI

You’ll learn how to set up Perplexity AI, configure search sources, prompt for deep research, and extract high‑quality, sourced answers quickly, saving time and improving research accuracy.

Takes
Text
Produces
Text
What you'll produce
Input
Explain the differences between GPU and TPU in machine learning.
Output
Answer: GPUs were originally designed for graphics rendering, focusing on parallel pixel calculations. TPUs, built by Google, are application‑specific integrated circuits optimized for tensor operations common in deep learning. GPUs excel at general‑purpose parallel tasks, while TPUs deliver higher throughput for matrix multiplications in neural networks. Sources: 1) Smith, A. (2023). "GPU vs TPU: Performance Benchmarks." Journal of AI Hardware. 2) Google AI Blog (2022). "Introducing TPUs."

The Workflow

8 steps · click a step number to mark it done

Open perplexity-ai and have it ready before you start

Create a Perplexity account or log in to access the dashboard.

Open Settings, go to the 'Personalization' tab, and add a brief instruction like 'Answer concisely with citations.'

In the main chat window, click the 'Sources' toggle and choose 'Academic' to limit results to scholarly papers.

Toggle on 'Deep Research' mode near the top of the chat so the model generates an in‑depth report.

Enter your research query (e.g., "Explain the differences between GPU and TPU in machine learning") and press Enter.

Read the structured answer; scroll to the 'Sources' panel at the bottom to view cited references.

Click the 'Export' button to copy the answer or download it as a PDF for your report.

To stay updated, click the three dots next to the chat, select 'New Task', enter the same query, set a daily cadence, and choose to receive the results via email.

Ready to run this workflow?

Open perplexity-ai and follow the steps above.

Open perplexity-ai
Did this workflow help?
1
🔍 Looking for AI tools? Try searching!