Perplexity Hub: Perplexity Guides for Computer, Comet, Credits & Workflows (2026)
Tested guides, credit strategies, and workflows for Perplexity Comet, Deep Research and Computer.
Last updated: August 2026. Actively maintained.
TL;DR: This page is Product with Attitude's independent, hands-on library of tested guides for Perplexity Search, Deep Research, Comet, Computer, and Personal Computer. I build and maintain it from my own projects. Each guide includes the full reality of my work: what I tried, what came out, how many credits it cost, and where it broke. There’s another ‘‘Perplexity Hub’’, an official one, at perplexity.ai/hub. But what you’re reading now is created, owned, and maintained by me.
I wrote my first Perplexity article in October 2025, not that long ago when you think about it. At the time, Comet was Perplexity’s headline act. Imperfect, certainly, but capable enough to earn the company a place on my watchlist.
Since then, Perplexity has added Computer, Personal Computer, Brain, Comet Plus for publishers, mobile apps, and become the tool I use daily. Many times.
I’ve been pushing it to its limits, breaking things on purpose, and wearing different hats in the process: testing it as a builder, product manager, writer, and beginner. Each perspective revealed a different strength and weakness.
This page collects everything I’ve published about Perplexity on Product with Attitude.
My credit strategies, step by step guides, workflow and release breakdowns, product manager’s commentary, competitor teardowns, and posts where Perplexity plays a supporting role alongside Claude and Replit.
A few things I want to make clear:
Every guide is tested on real projects.
The credit estimate comes from my own usage.
Every infographic you see started as a Miro board full of scattered thoughts.
I don’t skip the parts that don’t go as planned, and if you keep reading, neither will you.
What’s Inside
Part 2: Perplexity Computer
Part 4: Is Perplexity useful for your role? PMs, vibe coders, writers, and finance professionals.
FAQs
Hey, I’m Karo Zieminski,
AI Product Manager and builder.
I write Product with Attitude, an AI newsletter for tens of thousands of readers across 146 countries, helping them develop critical AI literacy the only way it sticks: through practice.
We don’t just use AI. We build workflows, agents, and products with it, while studying how AI itself is built, positioned, and woven into our work.
If you’re new here, welcome!
Part 1: What Is Perplexity, and Which Product Do You Need?
Perplexity is still best known as an AI-powered search and research platform, but I believe that framing undersells it. It’s becoming an ‘‘AI-everything’’ platform.
I use it for the whole rather crowded bench: research, design, builds, analysis, maintenance, deployment, documents, agents, task management, Github management, image generation and every tiny task in between.
As of August 2026, there are five core products:
Perplexity Search (the answer engine)
Comet (AI browser)
Perplexity Computer
Personal Computer
Perplexity API (Sonar)
The 5 Perplexity Products
Perplexity Search
Perplexity Search is for searching, as the name rather helpfully suggests. It’s the OG product and the main entry point for 45M+ monthly users. I use the standard Search only for simple queries: What’s the capital of Denmark? You ask, it answers, always with inline citations, so you can verify the source.
If I wanted to get an in-depth, sourced report on anything, that’s what I’d use. Deep Research. It scours the web, runs the findings through several leading AI models, synthesizes them into an answer, and adds inline citations so you can verify the sources.
I’ve never caught it hallucinating. Not even once. And I verify EVERYTHING.
To see where LLMs agree and where they give you different, equally confident answers, run your findings through Model Council. It’s fun.
Perplexity Search comes in three tiers: Free, Pro at $20/month, and Max at $200/month. I use Max because it gives me access to the full Perplexity lineup, including Computer.
Comet (AI browser)
My Perplexity story started with Comet, back when it was still the headline act. It's a Chromium-based AI browser with its own Assistant that you can use for multi-step in-browser tasks such as filling forms, or shopping for groceries. Since March 2026, it's free for everyone.
Perplexity Computer
This is the product I spend most of my month inside, so consider this section heavily tested and mildly biased.
Perplexity Computer is a cloud-based AI agent that orchestrates multiple frontier models (including Claude Fable 5, Claude Opus 4.8, GPT-5.6 Sol, Grok 4.5, GLM 5.2-based Preview, Gemini 3.1 Pro) to research, build, automate, and deploy from a single prompt.
The second half of this article focuses on Perplexity Computer guides.
Personal Computer
Perplexity Computer and Personal Computer sound almost identical. Easy mix-up.
Personal Computer = Perplexity Computer + Mac mini.
You get both the agentic layer and a device. Computer stays always on, follows your schedules and triggers, and can be controlled from your phone. I do exactly that, from some odd places where I have chosen, against all evidence of good sense, to get work done.
Perplexity API (Sonar)
This is the only product on this page I evaluate with my developer hat on. This is a B2D (business-to-developer) product: a full-stack, model-agnostic developer platform.
You can use it to add AI and internet-search capabilities to the apps you build. It includes four main tools:
Agent API: Lets AI complete multi-step tasks.
Search API: Finds current information on the internet.
Embeddings API: Helps software understand and compare the meaning of text.
Sandbox API: A secure environment where AI can run code and perform tasks safely.
It works across different AI models, so you avoid the usual single-provider handcuffs. It’s also compatible with OpenAI’s technical format.
Pricing: Pay-as-you-go at $1/million tokens.
Perplexity Products Comparison Table
Here’s how the five products compare: Perplexity Search (free to $200/month), Comet (free), Perplexity Computer (Pro and Max tiers), Personal Computer (Computer plus a Mac mini), and the Sonar API (pay-as-you-go at $1/million tokens):
Part 2: Perplexity Computer Guides
You got the definition in Part 1. This is the lived-in version. It's the agentic tier, and it's where the wow factor lives. It's also where most of my credits, testing hours, and an unreasonable share of my evenings go.
Computer unifies every current capability of AI into a single system.
—Aravind Srinivas
That quote, shared by Srinivas during the launch on February 25, 2026, may sound a touch overhyped, but it is accurate.
Computer can do what your Claude, Gemini, and ChatGPT can do, and I don’t mean that figuratively. It lets you choose whichever frontier model best suits the job. Or makes the choice for you when model selection is not how you wish to spend your afternoon.
Example: You’re building an app.
You work in Perplexity, but the actual build runs through Claude Code underneath, with Gemini supporting the design and ChatGPT checking your UX copy.
This orchestration, using the strongest parts of the best models for the right tasks, is why I am consistently more impressed by the results I get from Computer than by the results I get from each model separately.
But there’s more.
Perplexity Computer Brain
I’m going to write a dedicated article about Brain in the coming weeks. Until then, know that it launched on June 18, 2026, as a Research Preview for Max and Enterprise Max subscribers.
Brain is a self-improving memory system Perplexity Computer. It builds a working model of our projects, and files, then uses that context to push work forward.
I’ve had my fair share of setting up second brain systems. Notion databases with 47 views, Obsidian vaults with bidirectional links. Every time, it started with glorious clarity and ended in maintenance. The whole “second brain” paradigm - capture everything, retrieve it later - is a splendid idea, but it rarely mentions how much work it takes to maintain. A lot.
Perplexity Brain removes the system entirely. I don’t need to maintain anything. I carry on working, and it adapts to me.
Read more: Perplexity Computer vs ChatGPT Work vs Claude Cowork
What Can You Do and Build With Perplexity Computer?
These guides cover what Perplexity Computer is, what it costs, and how to use it without burning through too many credits.
Perplexity Computer Review: What I Built in One Night
I explain why Computer is novel, how it compares to Claude Cowork and OpenClaw, and what I built during my first night with it.
Perplexity Computer for Professional Finance
Perplexity Computer for Professional Finance launched May 4, 2026. I tested it on Palantir. The output was magnificent. From a Critical AI Literacy standpoint, the death of prompting worries me.
Perplexity Computer in May 2026: Updates That Changed Almost Every Serious Use Case
I shared five Perplexity Computer workflows I use every week, with the credit cost, model, time saved, PM read, and Critical AI Literacy check included for each.
How much does Perplexity Computer cost?
How to Save Credits on Perplexity Computer (Part 1)
I tested Computer hard to understand the pricing strategy and documented where credits disappear. This guide covers the credit system, a calibration experiment you can copy, and three workflow prompts I use to spend less.
Save More Credits: Advanced Patterns for Perplexity Computer (Part 2)
Seven techniques that cut credit consumption by 50%+. Covers the stacked-context tax, the fresh thread rule, splitting strategy from execution into separate threads, and when to replace conversation with scripts entirely. Part 2 of the credit-saving series.
Why Your Credits Vanish
I break down how I use Computer every single day, without smashing straight through the credit ceiling.
When Computer Wins (And When It Doesn’t)
Read the latest comparison: Perplexity Computer vs ChatGPT Work vs Claude Cowork
I’ve tested enough orchestrated workflows to have opinions with receipts.
We have five real options for orchestrated AI work today. Computer is one of them. Sending the wrong job to the wrong tool is the most expensive mistake we can make.
Where Does Perplexity Fit in Your AI Stack With Tools Like Claude or Replit?
Perplexity Computer vs ChatGPT Work vs Claude Cowork
What ChatGPT Work, Perplexity Computer and Claude Cowork assume about us, the users, and what they reveal about perceived AI intelligence.
10 Tools for Running a Bestselling Substack Newsletter in 2026
Perplexity is one of the core tools in my daily workflow alongside Claude, Recraft, and the rest. This post shows exactly how each tool fits.
Perplexity for Vibe Coders
A 16-step roadmap for building production-ready apps with AI. Perplexity serves as the research and fact-checking layer alongside Claude, Replit, and Cursor.
Part 3: Common Mistakes People Make With Perplexity
Most of these problems are habit-related. They have nothing to do with your skills.
Bringing all your Claude habits into Perplexity without adapting them was my initial mistake. Perplexity threads retain context. That’s one of the biggest differences between Computer and tools such as Claude Cowork. It also means your first prompt doesn’t have to be a complete brief that anticipates everything.
Not organizing threads into Spaces or Projects. The more I use Perplexity, the more useful I find this feature. It’s often positioned as a collaboration space, but I recommend exploring it even if you work alone.
Prompting like it’s still 2022. I wrote a dedicated article explaining how some prompting techniques can hurt the output of reasoning models.
Not prompting at all. Using Perplexity like Google. Don’t type in piles of keywords (“best AI newsletter tools 2025 comparison”, “critical AI literacy 2026”, that sort of thing). Perplexity will ask for more detail, so you’ll end up querying it multiple times and paying for every query.
Every answer comes with citations. Click them. Always verify the output. I mentioned that I’ve never caught Research hallucinating, but I can only say that because I verify everything.
Not spending enough time understanding how credits work. Computer credits burn fast. Use the bonus credits you receive at signup to learn what consumes them and how quickly.
Part 4: Perplexity for Your Role
Is Perplexity useful for product managers?
Oh, yes. The more I use Perplexity, the clearer it becomes that a product manager’s workflow maps almost directly onto its core capabilities.
Most PM work falls into three modes:
Finding out: Search, Research, and Academic help uncover market dynamics, customer problems, competitor moves, technical constraints, and evidence to support (or de-validate) your assumptions.
Deciding: Computer-built dashboards and synthesis inside Spaces pull patterns out of scattered information. You know, the material you need to make a decision instead of producing a VERY comprehensive pile of research.
Shipping the communication: Computer and connectors help translate decisions into briefs, reports, presentations, updates, and other stakeholder-ready outputs.
But here’s where PMs can get into trouble: when they outsource the judgment along with the research. Keep your product thinking close.
Is Perplexity useful for vibe coders?
At this point, it feels wrong to separate the PM role from the vibe-coder role, because 2026 has shown that they are one and the same.
The answer is yes. Perplexity is useful as the obvious vibe-coding tool and as a research and validation layer. The upfront research can save you hours of building beautiful nonsense. I also use it to manage my specs and GitHub repositories.
Is Perplexity useful for writers and newsletter operators?
I’m speaking from experience here, too. No other tool has helped me more with source discovery, fact-checking, article planning, competitive framing, and identifying adjacent questions my readers are already asking.
In theory, you can build custom research pipelines in Claude Code, and I’ve seen some writers do this with great success. I’m just not convinced that a solo build can realistically outperform the research features Perplexity has developed.
You can also use it to support your writing process, but hold on to your voice. And point of view. Otherwise, you’ll produce the same gray sludge everyone else is publishing.
Is Perplexity Computer useful for professional finance?
Yes. On May 4, 2026, Perplexity launched a dedicated feature: Perplexity Computer for Professional Finance. I documented my tests here.
Finance is a high-stakes use case, so treat outputs as analysis support. The investment decisions stay yours.
Part 5: About This Library
What is this Perplexity library?
It started with me running 45,000 credits of live testing across three weeks and documenting the workflows, the credits that vanished, and the workarounds that stopped the bleeding. Today, it's an independent collection of tested guides for Perplexity Search, Computer, Comet, and Apps. Every article comes from my own hands-on testing.
How is this library different from Perplexity’s own documentation?
I started this library with a Comet use-cases post in October 2025. Since then, it grew into one of the deepest independent Perplexity resources on Substack.
Perplexity explains what the products can do. I use between 23,000 and 70,000 credits for live testing each month, depending on how much I build. I retrace these credits. Write about decisions and consequences. Document every workflow and everything that breaks (or even stutters). If I had more time, I’d write a dozen more Perplexity Computer guides. The ones I do publish are the fossil record of my experiments. Nothing gets published here until it has survived a real project. The page you’re reading right now has been revised twelve times since the first version.
Who is this for?
I built this library thinking about the version of me who started with Perplexity in October 2025 and had no idea which product did what. Now it's the reference shelf I wish someone had handed me back then, for: writers, teachers, PMs, and builders. Free tier or a 10K-credit habit, there's something here.
Which article should I start with?
New to Computer? Start with the full review. If you’re already using it and burning credits, go straight to the credit-saving guides: Part 1 first, then Part 2 for the advanced techniques.
Frequently Asked Questions About Perplexity
What is Perplexity Search best for?
I use Search for quick queries and Deep Research when I need solid proof. Both features give me a synthesized answer with numbered citations. I click every one. If a source is an old Reddit from 2019, I downgrade it in my potential sources list. If it's a primary doc, I keep going.
What is Perplexity Comet?
I run Comet when the task lives inside the browser. Perplexity Comet is a browser-based AI assistant for productivity workflows. I gave it small jobs first, mostly to see what would break. It helps with tasks like content repurposing, and browser-context work.
Think of Comet as the layer for working inside the web, not just asking about the web. It becomes useful when the task requires navigation, comparison, summarization, or transformation across web pages and tabs.
What is Perplexity Computer?
This is the product this whole page orbits, and the one that eats most of my credits. Perplexity Computer is the agentic tier of Perplexity. It’s a cloud-based AI agent that can orchestrate frontier models to research, build, automate, deploy and maintain apps from a single prompt. I find it’s best for serious, scoped workflows where the output is valuable enough to justify agentic execution.
What can I build with Perplexity Computer?
You can use Perplexity Computer for multi-step workflows such as market research, financial analysis, product strategy, competitive teardowns, app prototyping, automation planning, and deployment-oriented tasks. I used it to build my homepage, mini apps or agents for myself, and products for my readers.
The strongest use cases I see are the ones with clear constraints. Computer can do powerful work, but it needs boundaries or it will happily send your credits up in smoke.
Some examples:
What does Perplexity Computer cost in 2026?
Perplexity Computer is available on the Pro plan at $20/month with 4,000 bonus credits and pay-per-use pricing, the Max plan at $200/month with 10,000 monthly credits and 35,000 bonus credits, and Enterprise Max at $325/seat/month.
Note: AI companies rearrange their pricing and packaging with astonishing enthusiasm. I last verified these prices in August 2026.
How do Perplexity Computer credits work? How to save them?
I learned the credit system the expensive way, which is why there are three articles below instead of one. In short, Perplexity Computer credits are consumed when Computer performs agentic work. I've documented all of it, credit by vanished credit, across several articles:
When should I use Academic focus in Perplexity?
Academic focus is my default for anything I expect readers to quote back at me. When I need peer-reviewed or research-grade sources. I lean on it hardest for AI literacy content. If readers may cite you later, the sourcing bar goes up.
WHY SUBSCRIBE ・YOUR BENEFITS・CRITICAL AI LITERACY・TOOLS I BUILT・CLAUDE HUB・PERPLEXITY HUB ・VIBE CODING HUB
Attitude Vault ・LinkSwap ・StackShelf
My free guides show which product does what and what it costs. The premium version is for readers who want advanced techniques, and the cost-saving patterns that can pay for themselves if they prevent one expensive agentic faceplant.
A very useful guide indeed Karo. The interesting shift is that these tools are becoming less about clever prompting and more about building a dependable personal workflow. That is where AI becomes infrastructure for me! Nice one.