Integrating DeepSeek with GoHighLevel

Welcome to this quick tutorial on how we're going to connect DeepSeek, which is the new Chinese AI, into GoHighLevel. I haven't seen any tutorials on this so far, so I wanted to make the first one because I think I just figured it out. So if you've gone to DeepSeek's website, one thing you'll notice is that on the top right-hand corner, they do have API access, which is great. But, as of a couple days ago, if you click on it, you don't have access to actually use their API. This is because, I think, one, their servers are overloaded because it's a very underfunded project. And two, I think they're under some malicious attacks, which is making it very difficult for people to access their API. So I'm going to show you how you can bypass that using a proxy and actually be able to connect this to GoHighLevel. So that's what we're going to be going through today, connecting DeepSeq to GoHighLevel. And the software we're going to be using, of course, is HighLevel, but specifically my version of HighLevel, which is Hexona Systems. It's just my white label. We're going to be using DeepSeq via something called Grok. And then we're also going to be using Make.com to make them connect to each other. This is the workflow that we're going to go through. So we're going to start off with the GoHighLevel webhook. We're going to send that to DeepSeek via Make via Grok, and I'm going to show you exactly what Grok is, and then back to GoHighLevel via Make. So whatever use case you want to do this for, that's completely up to you. I'm just going to show you an example of a chatbot, and then you can follow that. And again, I'm not going to go into the details of what DeepSeek is, why it's better than the other LLMs, or why you should choose it. I imagine that you already know that, and there's a million other tutorials that you can find on Google or YouTube about that. So this is just a practical application on exactly how you can use it. And real quick, if you don't know who I am, my name is Hamza Baig. I go by Hamza Automates. I run an award-winning seven-figure AI automation agency. I just won GoHeadLevel's SAS Per Neuro Platinum Award, putting me at the top 50 out of 60,000 agencies. I've also won their Affiliate Award, meaning that I'm in a very unique position. And of that, I haven't just done this for my own agency, but I've successfully been able to help others achieve the exact same results as well. So all that to say, if there's anyone you should listen to about Go High Level Automation AI, I'm your guy. So just jumping back into it, we're going to first start off with Grok. So I want you to go to Grok Cloud or just Google that. And you can basically sign up for it. It's like a compiler of all these different LLMs that you can access. But the cool thing is that you can access their API free of charge. And basically just generate your own API keys. So it's very simple. Go to it, sign up. You get access to it within minutes. You don't have to pay anything. and if you look around its playground here at the very top you do have access to deep seek's r1 model which is really cool and we can use that in our api now just jumping into go high level we're going to head over to the automation tab again i'm going to assume that you already know how to use go high level this is not a beginner's tutorial or a walkthrough so there's other videos that you can watch on learning the software i'm simply just showing you how to connect it to deep seek so what we're going to do is go to the automation tab i have this built up so i'm just going to walk you through it I use this for my chatbots, but you can basically use the premise for anything. The point is that we just want to send a webhook from GoHighLevel. So you can always do that by doing add action, search webhook, and then we're going to put our make.com URL there. So we're going to send whatever information we want from GoHighLevel and send that to make.com using a webhook. Now, the use case that I'm doing here is that I'm creating a chatbot. So the information that I'm sending via... A webhook to make is just a customer response. Whatever the customer messages me via SMS, I want to send that to make.com, and that's what I'm going to show you. Next, we're going to jump into make.com. So this is the actual workflow. It's very simple. The webhook is over here. If you know how webhooks work, basically we're sending something into the internet with an address on it, and then make is going to catch whatever we threw because we connected them via that address. So you can see here I've got the destination URL, which is the same destination URL as Make told me to put, so I've inputted that. So now the information has been received via Webhook. Then what we're going to do is send it to Grok, G-R-O-K, the platform that I just showed you. And then you're going to set up the Grok's module, connect your account to it, and then inside you can choose what model you want to work with. And of course, for this example, we're going to be using DeepSeek, so just select DeepSeek. And now just like any other LLM like chat GPT or anything else you would connect via make, you can use deep seek that way. So if you see here, you can fill out the user's role with the content of the message that came from the webhook. The assistant is where you put the prompt. So anything you want to train it, it is like a professional SMS agent for a restaurant. This is a very fictitious random example, but I just want to show you how it works in practice. You can put whatever you want inside this prompt. And then you've got the other parameters that you can change here, similar to how you would connect this to open AI. Then once you've outputted your response, you send it back to GoHighLevel using the update a contact function, or actually any function that you want. In the case of my chatbots, I'm using update a contact. But basically from the webhook, you want to recognize who you want to send this information back to. I put the phone number there as well, just to match it perfectly. And then I send DeepSeek's response back. to back to go high level the weird thing about deep seek and the nuance that it has is that it includes all of its thinking inside of the prompt inside of the output that it produces so i basically spent all night coming up with a way to just extract only the output and leave its its thoughts behind so that's what this whole mechanism is here but i'm going to give you access to this you can just copy paste it maybe i'll put in the description or inside my school community so you can access this and then you can extract just the output that you need you And that's basically how it's going to work. Webhook to DeepSeek via Grok to go high level. And now I want to show this to you in action. So I've already built up a chatbot because that's what I kind of do for a living. So I'm going to just message it and say, what specials do you have? And you can see here that the SMS came here via my phone number. So it's going to trigger my DeepSeek automation. So that just increased by one. And then we'll see here that this is going to update very shortly. If I go back. Oops, I hope it's saved. Yeah, we can see that it just ran right now. And then I should have a response that's come in. If you check the execution log. Yep, it sent the customer's reply to DeepSeek. And then my response has already come in. So I'm actually kind of late. So if you see here, hi to the specials of Ribeye, blah, blah, blah. Hope to see you soon. What time do you close? So I just said that. I just want to show you this in action. Yeah, so it's really quick. Like DeepSeek is fantastic. Like it's so quick. It's cheaper. And honestly, outputs aren't that bad either. So you can see here, this ran instantly. And my response should be coming in very soon. We close at 10 p.m. today. Let me know if you need anything. So the brain of this entire bot is set up by DeepSeek, which is awesome. So this is exactly how you set it up. And you can use this for a variety of different use cases. I'm going to share more examples. But this is how you connect both of them. So if you have any other follow-up questions about this automation specifically, you can always join my school community at school.com slash automate. The link will be in the description. But we just passed 22,000 members. I have a bunch of different tutorials in there, very similar to what you're watching right now. So if you want to get into agents and using Go high-level for AI automation specifically, that's what all of this is about. We have a fantastic number of wins there, so you can see people actually making a lot of money doing this. And DeepSeek is only going to help you improve your margins and offer you a more competitive... better product so you can always join that it's free link is in the description you can message me with any questions you have but yeah deep seek is awesome and keep building with it i'll see you inside

But, as of a couple days ago, if you click on it, you don't have access to actually use their API. This is because, I think, one, their servers are overloaded because it's a very underfunded project. And two, I think they're under some malicious attacks, which is making it very difficult for people to access their API.

So I'm going to show you how you can bypass that using a proxy and actually be able to connect this to GoHighLevel. So that's what we're going to be going through today, connecting DeepSeq to GoHighLevel. And the software we're going to be using, of course, is HighLevel, but specifically my version of HighLevel, which is Hexona Systems.

It's just my white label. We're going to be using DeepSeq via something called Grok. And then we're also going to be using Make.com to make them connect to each other.

This is the workflow that we're going to go through. So we're going to start off with the GoHighLevel webhook. We're going to send that to DeepSeek via Make via Grok, and I'm going to show you exactly what Grok is, and then back to GoHighLevel via Make. So whatever use case you want to do this for, that's completely up to you.

I'm just going to show you an example of a chatbot, and then you can follow that. And again, I'm not going to go into the details of what DeepSeek is, why it's better than the other LLMs, or why you should choose it. I imagine that you already know that, and there's a million other tutorials that you can find on Google or YouTube about that.

So this is just a practical application on exactly how you can use it. And real quick, if you don't know who I am, my name is Hamza Baig. I go by Hamza Automates.

I run an award-winning seven-figure AI automation agency. I just won GoHeadLevel's SAS Per Neuro Platinum Award, putting me at the top 50 out of 60,000 agencies. I've also won their Affiliate Award, meaning that I'm in a very unique position.

And of that, I haven't just done this for my own agency, but I've successfully been able to help others achieve the exact same results as well. So all that to say, if there's anyone you should listen to about Go High Level Automation AI, I'm your guy. So just jumping back into it, we're going to first start off with Grok. So I want you to go to Grok Cloud or just Google that.

And you can basically sign up for it. It's like a compiler of all these different LLMs that you can access. But the cool thing is that you can access their API free of charge. And basically just generate your own API keys. So it's very simple.

Go to it, sign up. You get access to it within minutes. You don't have to pay anything.

and if you look around its playground here at the very top you do have access to deep seek's r1 model which is really cool and we can use that in our api now just jumping into go high level we're going to head over to the automation tab again i'm going to assume that you already know how to use go high level this is not a beginner's tutorial or a walkthrough so there's other videos that you can watch on learning the software i'm simply just showing you how to connect it to deep seek so what we're going to do is go to the automation tab i have this built up so i'm just going to walk you through it I use this for my chatbots, but you can basically use the premise for anything. The point is that we just want to send a webhook from GoHighLevel. So you can always do that by doing add action, search webhook, and then we're going to put our make.com URL there. So we're going to send whatever information we want from GoHighLevel and send that to make.com using a webhook. Now, the use case that I'm doing here is that I'm creating a chatbot.

So the information that I'm sending via... A webhook to make is just a customer response. Whatever the customer messages me via SMS, I want to send that to make.com, and that's what I'm going to show you.

Next, we're going to jump into make.com. So this is the actual workflow. It's very simple. The webhook is over here. If you know how webhooks work, basically we're sending something into the internet with an address on it, and then make is going to catch whatever we threw because we connected them via that address.

So you can see here I've got the destination URL, which is the same destination URL as Make told me to put, so I've inputted that. So now the information has been received via Webhook. Then what we're going to do is send it to Grok, G-R-O-K, the platform that I just showed you. And then you're going to set up the Grok's module, connect your account to it, and then inside you can choose what model you want to work with. And of course, for this example, we're going to be using DeepSeek, so just select DeepSeek.

And now just like any other LLM like chat GPT or anything else you would connect via make, you can use deep seek that way. So if you see here, you can fill out the user's role with the content of the message that came from the webhook. The assistant is where you put the prompt. So anything you want to train it, it is like a professional SMS agent for a restaurant.

This is a very fictitious random example, but I just want to show you how it works in practice. You can put whatever you want inside this prompt. And then you've got the other parameters that you can change here, similar to how you would connect this to open AI. Then once you've outputted your response, you send it back to GoHighLevel using the update a contact function, or actually any function that you want.

In the case of my chatbots, I'm using update a contact. But basically from the webhook, you want to recognize who you want to send this information back to. I put the phone number there as well, just to match it perfectly. And then I send DeepSeek's response back. to back to go high level the weird thing about deep seek and the nuance that it has is that it includes all of its thinking inside of the prompt inside of the output that it produces so i basically spent all night coming up with a way to just extract only the output and leave its its thoughts behind so that's what this whole mechanism is here but i'm going to give you access to this you can just copy paste it maybe i'll put in the description or inside my school community so you can access this and then you can extract just the output that you need you And that's basically how it's going to work.

Webhook to DeepSeek via Grok to go high level. And now I want to show this to you in action. So I've already built up a chatbot because that's what I kind of do for a living. So I'm going to just message it and say, what specials do you have?

And you can see here that the SMS came here via my phone number. So it's going to trigger my DeepSeek automation. So that just increased by one.

And then we'll see here that this is going to update very shortly. If I go back. Oops, I hope it's saved. Yeah, we can see that it just ran right now.

And then I should have a response that's come in. If you check the execution log. Yep, it sent the customer's reply to DeepSeek. And then my response has already come in.

So I'm actually kind of late. So if you see here, hi to the specials of Ribeye, blah, blah, blah. Hope to see you soon.

What time do you close? So I just said that. I just want to show you this in action.

Yeah, so it's really quick. Like DeepSeek is fantastic. Like it's so quick.

It's cheaper. And honestly, outputs aren't that bad either. So you can see here, this ran instantly.

And my response should be coming in very soon. We close at 10 p.m. today.

Let me know if you need anything. So the brain of this entire bot is set up by DeepSeek, which is awesome. So this is exactly how you set it up. And you can use this for a variety of different use cases. I'm going to share more examples.

But this is how you connect both of them. So if you have any other follow-up questions about this automation specifically, you can always join my school community at school.com slash automate. The link will be in the description. But we just passed 22,000 members. I have a bunch of different tutorials in there, very similar to what you're watching right now.

So if you want to get into agents and using Go high-level for AI automation specifically, that's what all of this is about. We have a fantastic number of wins there, so you can see people actually making a lot of money doing this. And DeepSeek is only going to help you improve your margins and offer you a more competitive... better product so you can always join that it's free link is in the description you can message me with any questions you have but yeah deep seek is awesome and keep building with it i'll see you inside

Transcript for:Integrating DeepSeek with GoHighLevel

Transcript for:
Integrating DeepSeek with GoHighLevel