Introduction to Automation: ChatGPT 5 & N8n  ## Metadata - Author: [[YouTube]] - Full Title: Introduction to Automation: ChatGPT 5 & N8n - Category: #articles - Document Tags: [[video-notes]] [[youtube]] - Summary: Introduction to Automation: ChatGPT 5 & n8n - URL: https://www.youtube.com/watch?v=vDc_vxw8pvg Just a quick skim through overview of Chad GBT5 in case you're not familiar with it. It's smarter, has more contextaware responses, real time multimodal capabilities, agent mode and automation, expanded knowledge and fresh data access. So in open AI's new model, Chad GPT5, we have more than ever possibilities to create anything we want. What CAD GBT5 has to offer is improved reasoning, memory, and understanding of nuance for deeper and more accurate answers. So, one trend we'll see in the following uh pictures I'm about to provide is the accuracy of chat GPT5 as it increase from its predecessor models can also process and respond to text, images, and audio and code together for richer interactions. So instead of choosing a different modal you know for a specific prompt chatb5 is actually going to tune your prompt what type of information you give it and use the right modal within itself to give you the best answer. So now you don't have to go run to 4.5 4 to you know get a better answer in something. Chat GBT5 is the best expert at any question and any prompt. It also has built-in ability to create, manage and run anominous agents for complex workflows and multi-step tasks which today we actually are going to cover through the combination of open AI chatbt5 and N. Well, the question comes up about is how good is chat GBT5? How do we measure good? Can we compare it to other models? And yes, you can. So through the SWE bench verified benchmark test where a model gets its code repository and issue description then must create a working patch to fix the problem. Each test also notes how much reasoning the model has used. For fairness results exclude 23 out of 500 problems where our setup couldn't reliably check the solutions. In this test, GPT5 received a short prompt reminding it to carefully verify its fixes. This helped GPT5's performance but didn't improve 03's results. So from this we can gather that GT5 is more accurately giving us better answers compared to its predecessor models. So to add on to the discussion of how good is Chad GBT5 on this ader polyglot multi- language code editing chart GPT5 scored an impressive 88% compared to open AIO3 which had an 81 and GPT 4.1 which had a 52. So, GPT getting 88 on the Ader Polyglot benchmark test is so significant because it means GPT5 can accurately handle almost nine out of 10 realworld code editing tasks, setting a new record and showing a huge leap in reliability and usefulness for developers compared to any earlier AI model. So, for all my coders out there, Chad GBD5 just got way more accurate in coding and fixing your problems. And just to support the argument of how good Chad Gibbd5 is, um here's a bit more charts for you on how accurate it is compared to it other models and free form answers and multi-turn instruction following and API instruction following through internal OpenAI eval. So all across the boards, it's looking like Chad GBT5 is dominant within accuracy. And as per usual, you know, as we have mentioned before, this model seems to be the most accurate across all of them. Not just the past two, but all of them. So if we look at Chad GBT5, this little light pink, it's going to be right here cruising on the top, outperforming every single one of them. So as the input tokens increase, the accuracy obviously depletes in Chad GBD5, but it's still amongst the 90s to 100% whilst the other ones are all the way down here averaging in the 7060s. So what we can conclude from this is per usual, Chad GBD5 is going to be outperforming every pass model with insane increased accuracy. So what we can gather from this is Chad GBD5 is the most far accurate model we have ever received from OpenAI and the possibilities and the accuracy is endless. So why don't we bond this with automation and see what we can make. So how does Chad GBD5 and automation have any link together and how does it work? So through Chad GBD5 agents and N we can link the API and create whatever we want. So, CHGB5 introduces smarter automation. We have higher accuracy. So, we're going to get better answers, more accurate and more specific answers. It has all-in-one capability, so it can process text, images, code, and multi-step workflows. And it's for everyone. So, Nan's visual interface is very kind of simple, kind of easy to access. So, today's video I'm going to break down step-by-step guide on how to connect your chat GBT5 to your N and how to set up your first automation and also has a broad industry impact. So streamlines healthcare marketing, it is going to be pushing. This is a brand new kind of new thing that people are going to be diving into and it has lots of use. It has a lot of money into it and it's just awesome. It's great way to become more efficient and it just delivers a powerful more reliable automation for everyone reshaping how business and individuals work every day. So why don't we dive deep and see how Chad GP5 and N transform automation. And just to skim through real quick, Nan, in case you're not familiar with it, it's a automation platform which you can connect OpenAI to and other uh different APIs, open- source uh AIS. So today, this is going to be kind of the interface that we're going to be working with. Um you can link so many different things to it. You can create so many different automations out of it. Today, we're going to be covering kind of the basic Gmail send out Gmails automation that I'm going to show you how to do step by step. And after that, um, further down the line, I'm going to make more videos with N and OpenAI and showing you how to make more complex and more systematic, um, automation processes. So, real quick, what is N? It's an automation platform, has no code or lowode, has extensive integrations, flexible hosting, it's when you sign up, you get two free weeks, and then you have to pay. And it has scalable workflows. So you can have little small automations or you can have these large blown out automations that have so many little things to them and then the bottom line is N is allin-one automation hub linking your tools streamlining processes and saving time without heavy coding. So why don't we dive deep and see how this automation works. All right to begin we're going to jump over to N. Um, this is going to be our finished product, but I'm going to walk you through from the start step by step how to set it up and make sure it works. So, we're gonna click here. We're going to go workflow. We're going to go personal. Click save and this going to be a brand new project. Uh, we're going to click this plus and then we're going to scroll down to agent. We're going to click AI agent and then we're going to click out of off of it. And then boom. So, we've set up an AI agent in our automation. We have this kind of like big map we can work in and connect different things to. So uh we have our chat model, we have our memory, our tool and then I have our finishing process. So uh our chat model is going to be open AAI chatb5. Our memory is going to be just simple memory um that remembers five lines of code and then our tool is going to be obviously uh Google authorization so we can actually send out the emails um that we have made. So let's link our open AAI first things first. So if you click this little plus and then so search up open AAI chat model click that you have to make sure you link your open AAI account. So if you have not you don't know how to do that click create new credential and then you're going to get your API key from your API dashboard on open AI. You have do have to make sure that you have billing information on the AI API dashboard so everything links and actually works. Paste your API key here and then click save and everything should be fine. And then you're going to choose your modal. So, from here, we're going to choose GPT5-5, and it's going to be the latest one. And then after that, we can click off of it. And boom, we have added uh we're going to click memory. And then we're going to click simple memory. Simple as that. And then the context window length is going to be five. So, it's going to be perfect. So, boom. Now, we have a open AI chat modal and simple memory. If you guys want to try it, go to this chat bar down here and click Boom. So we said hello and it has responded to us. Hi, how can I help you today? And then it has pulled in from the memory. So we can actually see um right here save context and it's going to be around five. So how do we actually um make this work and be able to send out emails? Well, we have to connect the tool. So if we click tool and we're going to search up um Gmail tool. Click this. Boom. Just like that, we can connect our Gmail account through here. This is going to be similar to the API connection we had with OpenAI, except here, you're going to have to do it um a little differently. I did it through signing in through Google. Kind of gave me a little popup. So, it linked it. It's pretty easy. Um and after here, we can, you know, see what we're going to automate it with. So, it's going to be a message. It's going to send. And then this right here is going to be more of an expression. Um, two curly braces. And then we're click from AI. And then we're going to send here quotations email. And then if we go to subject, we're going to go expression and we're going to go um curly brackets. And then we're going to go from AI and then we're going to go quotations and we're going to say email body. This is actually going to be subject right here. So that's the subject. So So we have from AI email recipient subject and then body. Make sure you have it like this. And then we're going to click add option and click append and a attribution. We're going to turn this off so everything works properly. So we're going to click this off. We have that set up. Perfect. Now we have to configure our AI agent and how it sends the email. So you know we have you know who are we going to send the email to. Now we have to figure out how the email is going to look like. So if we click AI agent and click add option we do system message. This is going to be our kind of uh part where open AI comes in and helps us. Um we're going to go to open AI chat GP5. We're going to ask it to help us write an expression for this um system message that the AI agent reads before it sends an email. So, we're going to go to OpenAI real quick. Okay, coming over to OpenAI Chat GB5. I'm going to ask it to write me a kind of a code a uh instructions for my agent. And I pasted in a screenshot of my workflow. So, it kind of uh goes hand to hand and helps me out create the best version possible. So, it's actually writing a say an AI agent through a system prompt right here through its um SOP behavior and all of that. So, this is actually perfect. We don't need to worry about anything else but this right here. So, we're going to click copy. Come over here and we're going to click on our AI agent and we're going to go to expression. We're going to click this little button right here and then we're going to paste it in right here. Boom. Click out of it. And boom. Like that. After finishing everything up, we have our open AI chat modal. We have our memory, our tool, and then our output. We can go in the chat box and say, "Hey, send an email to saying that we have a at 4:00 p.m. on Thursday." Boom. We have our agent and memory that's running. And boom, just like that, we have on our output, your email has been sent to Daniel. I'll let you know if there's anything to reply. Boom. I've received an email from myself saying there's a business meeting scheduled for Thursday, 4 p.m. Please let me know if this time works for you, and I'd like to send you a calendar invite. Best regards. Just like that, we've set up an automation. We can use this to send emails to lots of people. You can click this tool and bind other things like databases of people's emails um other external things and you know make this way bigger than than what it is. So um nan is a perfect platform to bind with open AAI chat GBT5 to have successful automations and it's overall a complex but simple um kind of workflow. As you can see, this HUD isn't that complicated. Everything is kind of codef free. Everything you're doing is through chat GBD5 with its help and you're just kind of clicking buttons and connecting everything to have um a setup automation that's pretty straightforward. So yeah, um in today's video, we covered chat GB5's quick overview and how to do automation on N and overall how the workflow works. This is a very simple automation. uh by no means this is what automation is. Automation is way more complex and more, you know, has way more variables in it, but this is just a quick overview for you guys to um check out and see how everything works. So yeah, you guys enjoyed this video, please leave a like, leave a comment, and subscribe. If you guys want more N content and more AI tool content, let me know. I'll see you guys in the next one. Take care., Introduction to Automation: ChatGPT 5 & N8n, YouTube, 01k9gb4y9y5q39ymes7twtr9ht.