Resources · Make integration

Perplexity AI Make integrationAutomate Perplexity AI with Make.

Search the live web from inside an automation, then keep going. The Perplexity AI Make integration brings 6 modules, the bricks of your scenarios, the automations you build in Make. Five of them act, one searches. This page walks you from a blank canvas to a scenario that runs.

Verified Trustpilot reviews · AI, automation & growth agency

What you can automate

What can the Perplexity AI Make integration actually do?

It lets a scenario ask Perplexity AI a question and use the answer somewhere else. You feed it text from a form, a spreadsheet row or another app, Perplexity searches the web and writes back, and the next module carries that answer wherever it belongs. Six modules cover written answers, agent replies, text prepared for search, and the ranked results themselves.

Three shapes come back over and over. The first is the research brief: a list of topics sits in a spreadsheet, the scenario reads one row at a time, Create a chat completion asks Perplexity for a short summary, and the text goes straight back into the sheet.

The second is the sourced draft. List search results pulls the ranked pages Perplexity holds for a query, and the next module drops them into your writing tool or a team channel for review. The third is an internal library you can actually search: Generate embeddings turns each text into a run of comparable numbers, which is what lets a question find the right paragraph later on.

What Perplexity AI will not do inside Make is start anything. The app ships no trigger module, so a scenario always begins elsewhere: a schedule on the first module, or an event coming from another app. The doc publishes no size or rate limit for these modules, so the real ceiling is the credit balance on your account.

The same app also lives on n8n, and the Perplexity AI n8n integration covers it there. If the platform question is still open, n8n vs Make puts the criteria side by side. To turn a first scenario into a habit, Make training shortens the road.

Connect

How do you connect Perplexity AI to Make?

  1. 01

    Create your API key in Perplexity AI

    In your Perplexity AI account, open the settings from the Settings icon beside your name, then API in the left column. Create an API group, fill in your payment details, open the API Keys tab and hit + Create Key. That key is the secret Make will use to reach your account.

  2. 02

    Open the connection from Make

    Back in Make, drop any Perplexity AI module into your scenario and click Create a connection, the link between your account and Make that every module can reuse. The Connection name field is optional, though naming it after the account saves you a guess the day you run two of them.

  3. 03

    Paste the key and save

    Paste it into the API Key field, then click Save. If Make asks you to authenticate and confirm access, go ahead: that connection now works for every Perplexity AI module in every scenario you build.

First scenario

Your first scenario, one step at a time

GoalWhen the scenario runs, Make sends your query to Perplexity AI and brings back an answer built on the sources it found.

  1. 01

    Create the scenario

    From the Scenarios page, start a new one and click the + to place the first module. Search Perplexity AI by name and pick List search results.

  2. 02

    Link your account

    Click Create a connection, paste your API key, save. That connection stays available to every other Perplexity AI module you add to this scenario later.

  3. 03

    Write the query

    Fill in the query you want searched. For a first run, keep it narrow, a single topic you could check yourself in a couple of minutes.

  4. 04

    Add the second module

    Click the + to the right of the search module and add Create a chat completion. Map the output of the first into the input of the second so the model writes from what Perplexity found.

  5. 05

    Test with Run once

    Click Run once. The scenario executes a single time and each module shows the bundles it received, a bundle being one item travelling from one module to the next.

  6. 06

    Schedule it, then switch it on

    Open the clock on the first module, choose how often it should run, then flip the scenario switch. A brand new scenario sits at 15 minutes until you change it.

Modules

What each Perplexity AI module does

Perplexity AI gives you 6 modules in 2 groups. For each one: what it does for you, when to reach for it, and what to watch out for.

Actions

1 module

Make an API CallFor an example API call, refer to the Perplexity AI documentation.

Perplexity AI1

Create a chat completion

Action

In MakeCreates a chat completions with the chosen model.

Send an instruction and get written text back from the Perplexity model you select in the module, the version that actually writes the answer. The answer arrives as plain text, ready for the next module to store it, send it or publish it.

When to use it
a topic list where every row should come back as three readable sentences.
Watch out
each run spends credits on your Perplexity AI account, so test on one row before you point the scenario at the whole list.

Other modules

5 modules
Perplexity AI2

Create an agent response

Action

In MakeGenerate a response for the provided input with optional web search and reasoning.

Hand the module an input and it works the answer out, with web search and reasoning switched on or off depending on your settings. What comes back is considered rather than fired off in one pass.

When to use it
a question arriving through a form that has to be checked against current sources before anyone replies.
Watch out
reasoning and search stretch the run, which matters on the Free plan where a scenario stops after 5 minutes.
Perplexity AI3

Generate contextualized embeddings (advanced)

Action

In MakeGenerate contextualized embeddings for document chunks. Chunks from the same document share context awareness, improving retrieval quality for document-based applications.

This one turns the pieces of a single document into embeddings, the runs of numbers a machine uses to compare meanings. Pieces cut from the same file keep the context of their neighbours, so a later search lands on the right passage instead of one that merely looks related.

When to use it
a contract or a handbook sliced into sections you want to query as one whole.
Watch out
the advanced version only pays off when the pieces belong together, unrelated texts gain nothing from it.
Perplexity AI4

Generate embeddings

Action

In MakeGenerate vector representations (embeddings) for the text you provide.

Give it text, get the vector form of that text back: a list of numbers standing in for the meaning, ready to be stored and compared. This is the plain version, one text at a time, with no shared context.

When to use it
sorting support tickets as they land so the similar ones end up near each other.
Watch out
those numbers are useless on their own, so decide where they go next, a database for instance, before you build the scenario.
Perplexity AI5

Make an API Call

Action

In MakePerforms an arbitrary authorized API call.

Call any endpoint of the Perplexity AI API straight from the scenario, using the connection already in place. Whatever the five ready-made modules leave out still goes through here.

When to use it
a feature Perplexity has just shipped that has no module of its own yet.
Watch out
nothing is prefilled, you write the call yourself with the Perplexity AI documentation open, so keep it for what the other modules cannot reach.
Perplexity AI6

List search results

Search

In MakeReturns the ranked search results from Perplexity's continously refreshed index.

Query the index Perplexity keeps refreshed and the ranked results come back, each one as a bundle travelling to the next module. You end up with the sources, not with a written answer.

When to use it
a monitoring scenario that gathers what gets published on a topic and files the links somewhere readable.
Watch out
a search can return nothing, one item or several, and the modules after it run once per item, each costing an operation, the unit Make counts against your plan.
Need help

Need help automating Perplexity AI with Make?

A person reads every message.

FAQ

Questions that come up next

01Is the Perplexity AI Make integration free?
Yes on the Make side. Perplexity AI is a standard app, available from the Free plan like the 3000 or so other standard apps in the catalogue, so the connector itself costs you nothing extra. The Free plan carries 2 active scenarios, leaves at least 15 minutes between two scheduled runs, cuts a run off after 5 minutes, and caps files at 5 MB with 512 MB of transfer. The other side is not free: Perplexity bills API calls against the credits on your account, and every module run spends some.
02What do you need to connect Perplexity AI to Make?
A Perplexity AI account and an API key with credits behind it. The key comes from the API portal of your account: open the settings, go to API, create an API group, enter your payment details, then generate the key from the API Keys tab and copy it. In Make, add any Perplexity AI module, click Create a connection, paste the key and save. Entering payment details does not charge you at that point; billing follows the calls you make afterwards. Nothing else is needed, no app to install, no admin rights on Make.
03Does Perplexity AI trigger a Make scenario in real time?
No. The app carries no trigger module at all, so nothing on the Perplexity side can wake Make up. Your scenario starts somewhere else: with a schedule on the first module, where Make runs the chain at the interval you choose, 15 minutes apart at the tightest on the Free plan; or with another app's trigger, a new row, a submitted form, an incoming message, with the Perplexity module placed behind it.
04What if a Perplexity AI module is missing in Make?
Reach for Make an API Call. It is one of the six modules and it hits any endpoint of the Perplexity AI API through the connection you already set up, so a feature without a module of its own is still within reach of the scenario. You write the call yourself from the Perplexity AI documentation, more work than filling a form but everything stays in one scenario. And if that still does not fit, another app in your chain often handles the job, since search plus completion already cover most of the reading and writing.
05Make or n8n for Perplexity AI?
It depends on who will maintain the automation, not on the app: both platforms reach Perplexity AI. Make works on a visual canvas where you drop modules and wire them together, and after a Run once you can read what each module received, which helps while you are learning. n8n is the other road, and the comparison page on this site runs through the criteria without crowning a winner. The useful question is human: pick the tool that person will be able to open, read and fix six months from now.
Hack'celeration Lab

Get our weekly automation tips.

No spam. Unsubscribe anytime.