Access Perplexity API In n8n To Scrape Live Website Data Through HTTP Request Authorization 🤖

Summary

In this video, Josh Pook demonstrates how to integrate Perplexity's API into n8n, an open-source automation tool similar to Zapier and Make.com. He guides viewers through setting up an HTTP node in n8n to perform a POST request for chat completions with Perplexity, explaining the process of adding headers and body parameters. Josh also highlights the benefits of using Perplexity for real-time data and research in automation workflows, comparing it favorably to other AI APIs. The video concludes with a test of the setup and a note on the current limitations of accessing citations and images through the API.

Takeaways

• 😀 Josh Pook is the presenter of the video, focusing on integrating Perplexity's API into n8n for automation sequences.
• 🤖 n8n is an open-source alternative to Zapier and Make.com for automation, which can be self-hosted to avoid costs.
• 🔗 There is no native Perplexity node in n8n, but it can be integrated using HTTP requests.
• 📚 The main action available with Perplexity's API is 'chat completions', which is straightforward to implement.
• 🌐 The API endpoint for chat completions is provided, and a POST HTTP request is used for the integration.
• 🔑 Authentication for the API is done using a header with a Bearer token, which requires an API key from the Perplexity account.
• 📝 The video includes instructions on setting up the HTTP node in n8n with the correct headers and body for the request.
• 📈 Perplexity offers different models with varying parameters and capabilities, including the ability to perform online searches.
• 💰 The pricing for using Perplexity's API is discussed, with a cost of $5 per thousand requests for the 'large' model.
• 🚫 Currently, features like citations and images are in a closed beta and not fully available to all users.
• 🔍 The video demonstrates a test of the integration, showing how to ask questions and request data from Perplexity.

Q & A

What is the main purpose of the video by Josh Pook?

The main purpose of the video is to demonstrate how to connect Perplexity's API into n8n node modules and scenarios for automation sequences.

What is n8n and how does it relate to Perplexity?

n8n is an open-source automation workflow tool, similar to Zapier and Make.com, which can be self-hosted. The video shows how to integrate Perplexity's API into n8n despite n8n not having a native Perplexity node.

Why would someone choose to self-host n8n?

Self-hosting n8n is beneficial because it is open source, which means there are no costs associated with using it, making it ideal for those with extensive automation needs.

What is the primary action available with Perplexity's API according to the video?

The primary action available with Perplexity's API is 'chat completions', which is used for generating responses based on input prompts.

How does one make an API call to Perplexity's API for chat completions?

An API call to Perplexity's chat completions is made using a POST HTTP request to the provided URL with the necessary headers and JSON body parameters.

What are some of the body parameters that can be included in a request to Perplexity's API?

Some of the body parameters include 'max_tokens', 'temperature', 'top_p', 'return_citation', and 'return_images', which allow customization of the API response.

What is the significance of the 'return_citations' and 'return_images' parameters in the API request?

The 'return_citations' and 'return_images' parameters, when included, instruct the API to provide citations and images along with the chat completion response, adding depth to the information provided.

How can one obtain the API key needed for authentication with Perplexity's API?

To obtain the API key, one must go to their Perplexity account settings, navigate to the API section, ensure they have credits, and then generate an API key.

What are the different models supported by Perplexity's API as mentioned in the video?

The supported models include 'llama 3', 'sonar small 32k', 'chat 32k', 'online 32k', and models with 7 billion and 8 billion parameters.

What is the current status of accessing citations and images through Perplexity's API?

As of the video's recording, accessing citations and images through Perplexity's API is limited and in closed beta, with access granted to users who apply and describe their use case.

How does Perplexity's API differ from other AI APIs like OpenAI's in terms of research capabilities?

Perplexity's API is noted for its ability to perform online searches and provide live, accurate data, making it superior for research purposes compared to other AI APIs like OpenAI's.

Outlines

00:00

🤖 Introducing Perplexity API Integration with n8n

Josh Pook introduces a tutorial on integrating the Perplexity API into n8n, an open-source automation tool similar to Zapier and Make.com. He highlights the benefits of self-hosting n8n and mentions that Perplexity does not have a native n8n node. The video will guide viewers on setting up an HTTP request to utilize Perplexity's chat completion feature within n8n workflows. The process involves using the API endpoint for chat completions, setting up headers for JSON acceptance and content type, and crafting the JSON body for the request.

05:01

🔗 Configuring Perplexity API with n8n HTTP Node

The tutorial continues with a step-by-step guide on configuring the Perplexity API within the n8n HTTP node. It includes creating a generic credential for the API key, setting request headers to accept JSON and specify the content type as application/json. The JSON body is then detailed, with parameters such as 'model' and 'messages' that can be customized according to the API documentation. The video also touches on the possibility of using different models and the option to enable online search capabilities for more accurate and up-to-date information.

10:04

🔍 Testing Perplexity API with Citations and Images

In the final part of the script, Josh tests the Perplexity API integration by sending a request to find out an interesting news event from the previous day, with the expectation of receiving citations and images. However, upon investigation, it's revealed that access to citations and images is currently in closed beta and not fully available to all users. The feature requires users to apply for access by filling out a form. The video concludes with a reminder about the benefits of using Perplexity for research and the potential of its API for obtaining live and accurate data, differentiating it from other AI APIs like OpenAI.

Keywords

💡n8n

n8n is an open-source automation workflow tool that serves as an alternative to platforms like Zapier and Make.com. It allows users to create automated sequences for various tasks without the need for coding. In the video, the host discusses how to integrate Perplexity's API into n8n for enhanced functionality, such as search and research capabilities within automation sequences.

💡Perplexity API

The Perplexity API is a tool that provides advanced search and chat completion functionalities. It is highlighted in the video as a service that can be integrated into n8n to add intelligent search capabilities to automation workflows. The API is demonstrated to be used for tasks like retrieving current news and information, with the potential for citation and image retrieval.

💡HTTP request

An HTTP request is a message sent from a client to a server to request access to a resource. In the context of the video, the host explains how to set up an HTTP request within n8n to communicate with the Perplexity API, specifically using a POST method to send data and receive responses.

💡API key

An API key is a unique code that identifies a user or calling program to an API. In the script, the host mentions generating an API key from the Perplexity account, which is then used to authenticate requests to the Perplexity API within the n8n automation workflow.

💡JSON

JSON (JavaScript Object Notation) is a lightweight data-interchange format that is easy for humans to read and write and for machines to parse and generate. The video script describes using JSON to structure the body of the HTTP request sent to the Perplexity API, including parameters for customization like 'max tokens' and 'temperature'.

💡Automation sequences

Automation sequences refer to a series of actions or steps that are automatically executed in a specific order. The video's main theme revolves around setting up such sequences in n8n, enhanced by integrating the Perplexity API to perform tasks like searching and retrieving information.

💡Research builds

Research builds in the context of the video refer to the creation of automated workflows designed for gathering and processing information for research purposes. The integration of the Perplexity API into n8n is positioned as a way to improve these builds by providing up-to-date and accurate data.

💡Rate limits

Rate limits are restrictions put in place by APIs to control the amount of data or number of requests a user can make within a certain time frame. The script mentions checking rate limits for the Perplexity API, indicating an awareness of these constraints when building automation workflows.

💡Pricing tier

A pricing tier refers to the different levels of service or access offered by a product or service at varying costs. The video mentions n8n having a pricing tier for those who want the service to be hosted, with self-hosting being an open-source and cost-free alternative.

💡Citations

Citations are references to the sources of information used in a piece of work. In the video, the host expresses interest in the Perplexity API's ability to return citations, which would provide the origins of the information retrieved, although it is later clarified that this feature is in beta and not fully available.

💡Images

The term 'images' in the script refers to the potential for the Perplexity API to return visual data along with the information requested. The host demonstrates setting up the API request to include image retrieval, although it is noted that this feature is also part of the beta program and not yet widely accessible.

Highlights

Josh Pook demonstrates how to integrate Perplexity's API into n8n for automation sequences.

n8n is an open-source alternative to Zapier and Make.com for self-hosted automation.

n8n lacks a native Perplexity node, but integration is possible through custom methods.

Make.com offers a native Perplexity module with options for chat completion or API calls.

The Perplexity API primarily supports chat completions with a single action.

An HTTP POST request is used to interact with the Perplexity API at a specific endpoint.

Example body parameters for the API request are detailed, including max tokens and temperature settings.

The video includes a tutorial on setting up an HTTP node in n8n for API requests.

Instructions on creating and using API keys for authentication with Perplexity are provided.

Headers for the API request include 'Accept' and 'Content-Type' set to 'application/json'.

The body of the API request is formatted in JSON with specific parameters detailed.

Different models of Perplexity are discussed, including the large parameter and online models.

The video mentions the capability of the online model to perform searches as part of its responses.

Rate limits and pricing for the Perplexity API are discussed, highlighting the cost per thousand requests.

The video tests the setup by asking a question and expecting a precise and concise answer with citations and images.

Citations and images through the Perplexity API are currently in closed beta and not fully available.

The video concludes with a mention of the author's book 'Client Acquisition Code' and a call to join the STRI community.

Transcripts

what's going on everybody Josh pook here

and in today's video I'm going to show

you how you can easily connect

perplexities API into your

n8n node modules and scenarios so you

can actually use perplexity to shoot

search the rep do research Etc when

you're building out your automation

sequences in NN you don't have to worry

about that Na and doesn't have a native

perplexity node just like how make does

make.com or you can do this All In na

we're going to dive into it let's Dive

Right

[Music]

In All right so really quickly before I

get in if you're not familiar with n8n

it's an alternative to zapier and

make.com yes there is a pricing tier

here if you want them to host it I

personally self-host it and once you

self-host it it is open source so you

don't have to pay

for it which is one of the benefits and

if you're using you know a lot of

automations it's definitely something I

would recommend looking into I did about

three videos the last couple days um the

main one I would suggest on checking out

is showing you how to host it on your

own VPS all right giving you the exact

command prompts Etc it's a bit longer of

a video it's about 30 40 minutes um but

it's going to show you how to get set up

so you don't have to worry about it in

the future um make.com I love still

using make.com but um yeah this is you

know here's the pricing here and if you

go into

make.com the good thing is they do have

a native perplexity module so if I go

like this for example perplexity you

really have two options you can create a

chat completion or make an API call

which is literally the same pretty much

just as an

HTTP um and yeah so that is the same

thing that we're going to be doing in

n8n so I'll leave the uh perplexity um

documents in the description below so

you can actually go and see them as well

but essentially there's really a few

things that we're going to look at here

so right here this is the API reference

and there is literally only one um

action that you can do which is chat

completions all right so it makes it

very simple all we're going to be doing

is a post HTTP request to this link

right here https

corand api. perplexity a/ chat for/

completions all right um you'll see some

example body parameters here different

things you can add here um objects you

can add different things like Max tokens

temperature top P return citation o this

is

cool so this is actually cool I like

this return

citations uh return images this is cool

I didn't know you could do this um

stream determines server so there's

different options you can potentially

use and if you go over here to the

language you can see like the shell so

you get the Cur request here if you want

to do this on the shell you can see the

headers right here so we're going to use

a header that is an accept Json uh so

we're just accept Json application Json

and then content type application Json

and then once we do that we're just

going to go to the Json right here and

add this into our body within Ann so

it's very simple I'm going to show you

exactly how to do it from scratch so

just for testing purposes I'm just uh

putting the trigger as in N when

clicking test workflow so once we click

test workflow we'd be able to see this

and what you're going to do is you're

going to add an HTTP node so I'll show

you what that looks like we would just

go here and click

HTTP we do an HTTP request and we put it

to a post all right then we would go

back to

our perplexity

documentation and we are going to grab

this URL right here or chat

completions and we're going to paste

that URL here authentification so um we

will use generic credential

type um and then for your generic off

type you're going to use a header off

and then um I could select my perplexity

authorization that I already have just

for purposes of this video I'll show you

what it looks like so you would go

create new you can change your name here

um here you would just put

off

authorization in the name and then in

the value you're just going to put

be like that oh wait you can't see so

you're going to write

bear like that and then you're going to

do a space and then you're going to go

over to your perplexity account you

would have to go here and uh or go click

on the the gear icon to go to your

settings then you click on

API make sure you have some credits uh

uploaded um so you're going to have to

buy credits I think it gives you $5 of

free credits every month you should be

good then you're going to generate an

API key these are hidden right now just

for the purposes of this video you're

going to copy that API key

and then when you're over here say if

this was my API key like this I would

just take this so it's bare with a space

in between then your API key and you

would just paste that for the value and

then you would click save okay so I'm

not going to do that we're going to

close and I'm just going to use my my

actual perplexity

authorization okay next we are going to

send some headers like I showed you so

we're going to come here we're going to

go here and we're going to grab

accept come here paste in that first

name come back and grab Json

application paste that in add another

parameter same value right

there and content type right

here boom okay and then we're going to

select send

body and we're going to be sending

Json and specify body type is going to

be using

Json and then we're just going to come

to Json here and you could just copy

this so if you make changes over on this

side like if I added a top P right here

um maximum is one if I added a top P you

can see that it actually shows over here

but I'm not going to do any of that you

could play around and add different

stuff here and it's going to show up

over here in the Json we're just going

to copy

this and we're going to come over here

and paste this into the

body like so now a few different things

you could change the model and if you go

over into the API docs and then you

click on guides and click on supported

models you'll see the supported models

so we got the perplexity models like

llama 3 sonar small 32k chat 32k online

32k and then the 7 billion parameter so

we have a 70 billion parameter and the 8

billion parameter um

32k and online and chat so we'll go

ahead if I'm not already using this one

we'll use this the biggest

parameter yeah so we'll use the large

parameter and the online so online

allows it of course so you can um you

leverage search you this will be really

good for research sh research builds and

stuff like that which I'll show you more

um in the future on this channel so make

sure to subscribe to stay tuned for

those builds um and then there are open

source models too so if you want to use

these hugging face ones like llama

38b instruct 7 uh 370b instruct and

87b instruct contacts lengths show right

here all right so I guess for the online

for this one so actually it's not that

bad so the online does have a bit

smaller contacts length but not that bad

and let's just see rate limits you can

check out some of these uh more

information here about rate

limits we got right here as well as

pricing so you can see the different

pricing here we're at the large so we're

the same as the small five bucks per

thousand uh requests so pretty solid

pricing and this allows you of course

like I said to do online searching

through the API which is useful you

can't really do this uh at this quality

of you know uh research with things like

chat GPT a through the API and

perplexity is just the best I've seen

for researching in general right so here

we could set you could set as many as

you want we're going to set you know you

can set your system role and your your

message this one for example is just be

precise and con uh precise and concise

and then role user how many stars are

there in the galaxy we could change this

to like

um what

is

um something

interesting that

happened

yesterday in the news so this will show

and um tell me

what date and time it

happened with sources so let's see how

this does so we're just basically

telling it to be precise and concise

asking this question and we want it to

return citations and images so let's go

ahead and test

this executing so we got it back let's

check it out um yesterday on June 21st

2024 Taylor Swift reunited with Prince

William at the era tour in London

all right

um

okay one thing I'm curious about is

where we would actually see the

citations and

the okay so I just searched perplexity

can I access citations and images when

accessing the perplexity API So

currently citations and images through

perplexity API is limited and not fully

available to all users here are some key

points uh regarding these features close

beta so basically it's saying is close

en Clos beta uh users interested in this

feature need to apply for Access by

filling in a form and describing their

use

case yeah so essentially unless you know

unless it's been recently updated unless

this is wrong guess you can't access the

citations and images yet even though it

says that here I guess it's just CU it's

in beta um oh yeah if you if if I if

maybe if you're watching this in the

future you can at this point or maybe

I'm just missing something here so if

you if I am let me know in the comments

down below but from what I see here as

well as what perplexity is saying um

it's saying that you can't do it yet

unless you're in the beta program which

I think would be nice I may even apply

through that form to see if I can get

access to the citations because I like

knowing where this information comes

from uh I like to fact check if I'm

leveraging things as well as to like uh

just nice to know um in general but

either way this is a lot better than

when you're doing re research um than

just using the open AI API because you

can get live data accurate data and um

really good stuff so other than that

guys I hope you got some value from this

video if you did make sure to drop a

like below and if you're new to this

channel make sure to subscribe we upload

videos every single day on AI news AI

tutorials how to leverage automation

make.com

n excuse me go highle marketing Etc how

to grow your business and also too if

you haven't already joined our free stri

community sh community.com link for that

will be down below we have about 750

members I believe in our Facebook and

Discord channel so Facebook group and

then also to guys check out my book

client acquisition code it's an in-depth

book showing you how to get clients

leveraging AI for your business if

you're in B2B if you're uh AI Enthusiast

uh entrepreneur Etc marketing agency

coach consultant definitely want to

check out that book link for that will

be in down below other than that guys I

will see you in tomorrow's video keep

hustling keep grinding and of course

accelerate your stride take care