How to Build Sora 2 Videos in n8n (NO CODE Tutorial)

n8nFresh📅 2025-11-01

Watch on YouTube ↗

Build Process

┌─────────────────────────────────────────────────────────────────────┐
│                    BUILD APPLICATION WORKFLOW                            │
├─────────────────────────────────────────────────────────────────────┤
│                                                                     │
│   PLANNING                                                          │
│   ────────                                                          │
│   ┌─────────────┐    ┌─────────────┐    ┌─────────────┐            │
│   │   Define    │───►│   Choose    │───►│   Setup     │            │
│   │   Scope     │    │   Stack     │    │   Project   │            │
│   └─────────────┘    └─────────────┘    └─────────────┘            │
│                                               │                     │
│                                               ▼                     │
│   DEVELOPMENT                                                       │
│   ───────────                                                       │
│   ┌─────────────┐    ┌─────────────┐    ┌─────────────┐            │
│   │   Build     │───►│   Add AI    │───►│   Test &    │            │
│   │   Features  │    │   Logic     │    │   Debug     │            │
│   └─────────────┘    └─────────────┘    └─────────────┘            │
│                                               │                     │
│                                               ▼                     │
│   DEPLOYMENT                                                        │
│   ──────────                                                        │
│   ┌─────────────┐    ┌─────────────┐                               │
│   │   Deploy    │───►│   Application   │ ◄── LIVE!                 │
│   │   to Cloud  │    │   Running   │                               │
│   └─────────────┘    └─────────────┘                               │
│                                                                     │
└─────────────────────────────────────────────────────────────────────┘

Key Components

• Project architecture
• Core features implementation
• Best practices

Transcript

[00:00] In this video, I'll show you how to build an AI video generator using Sora 2 and N8N. And N8N just released this note that makes it incredibly easy. If you're new to Sora 2, it's a new video generation model from OpenAI, and this model creates incredibly detailed videos, and these also include audio, like voice, music, and sound effects. This makes Sora 2 an incredibly popular model to use in marketing and advertising, social media content creation, like

[00:30] faceless YouTube channels, e-commerce and product visualization, educational content, and a lot, lot more. Unfortunately, getting access to Sora can be quite tricky. You need an invite code, and even then, Sora is only available in certain countries. But thankfully, we can use Sora 2 in N8N to build our very own AI video generators. And in this video, I'll show you how. This video is beginner-friendly, so if you really know how to use N8N, then skip ahead to the section where we

[01:02] discuss the video generation node. I'll also show you prompting tips and tricks for getting the best results out of Sora 2. So let's jump in. If you're new to N8N, it's a fantastic platform for building complex workflows without needing to write any code. All you have to do is drag and drop nodes onto your canvas, and you can build some very cool workflows. There are also many different ways in which you can access N8N. Some of them free, and others are paid for. And of course, there

[01:32] are pros and cons to each. In order to use N8N for free on your own machine, simply open up your terminal or command prompt, and enter the command npm install g n8n. If this command fails, you do need to install node.js, so simply go to node.js.org and install node.js. After installation completes, you can start N8N with the command npx n8n start. N8N will then give you this URL, and if you click on that URL,

[02:02] you can start using N8N right away. Running N8N on your own machine could be perfect for your use case. But of course, if you want to access N8N from anywhere, you will need to run N8N in the cloud or on a VPS. The simplest solution is to simply sign up for N8N cloud, and you can try out N8N for free using the link in the description. But if you have a look at the pricing page, you can see that N8N cloud will cost you about 24 euros per month. Now this is not cheap, but the N8N cloud version does include a

[02:33] lot of advanced features that's simply not available in the free community version. Now you can try out all those additional features by using the trial period, and again, if you use the link in the description, you will be supporting my channel. But I'm actually going to show you an even cheaper solution. You can save a lot of money by simply self-hosting N8N on your own virtual private server. The cheapest and easiest solution I found is to host N8N with Hostinger. They are sponsoring this video, but I've been using Hostinger

[03:04] since long before they became a sponsor. And with the current Black Friday special, you can host N8N for as little as $5 per month. So use the link in the description to go through this page. Then simply select your plan, like I'll go with this KMV2 plan. Then from this page, select your period, like one month, 12 months, or 24 months. Then scrolling down, click on application and N8N. You can get an additional 10% off if you use the coupon code Leon. And there we go.

[03:35] Over and above the Black Friday deal, we now also get an additional 10% off. After you've completed checkout, you should see a dashboard similar to this. All you have to do is click on manage app. Then let's complete this form. Then I'm going to close this pop-up. And if you see this pop-up, definitely send yourself a license key. Then go to settings, click on enter activation key. Then copy the license key from that email, add it to N8N and activate. Done.

[04:05] Let's go back to the dashboard. So whether you're running N8N locally or using the N8N cloud service or self-hosting it, you can now follow along with this video. The first thing we want to do is create our new workflow. Then let's rename it to something like Sora 2 demo. All right. So how do we generate Sora 2 videos? Well, what we have to do is click on add step. And from here, we need to add a trigger node. We can use triggers to fire off this workflow. Now we could trigger the workflow manually or maybe

[04:36] when some event happens. For instance, we receive a form submission or an email. Or maybe we want to run this workflow on a schedule using the on schedule trigger. Let's go with the form submission. For the form title, let's call this video generator. And for the form elements, let's call this field prompt. And for the type, let's select a text area. This is indeed a required field. And when we go back to the canvas, we can execute this workflow to give it a spin.

[05:06] Great. Let's give it a prompt like a ballerina dancing on the beach when a dog suddenly starts chasing her. Let's submit this. Cool. Now that we've submitted the form, we can see that this node is green, which means it was successfully executed. And when we click on the node, we can see its outputs on the right hand side over here. So here we can see the content of a prompt field, which was a ballerina dancing on the beach. Right. Now we want to pass this prompt to this video generation model. So let's click on

[05:36] this add a action button. And under AI, let's select the open AI node. Then scrolling down with this list, let's select generate a video. Now we can see the output from the previous node on the left hand side. Then in the middle, we can see all the parameters for this node. For instance, we can rename the note by clicking on this title. Let's call this Sora 2. Then we have to provide certain information like the credentials and the model name. Now for the credentials, let's click on

[06:08] create a new credential. And on the screen, we have to provide an API key. To get your API key, go to platform.openai.com slash API keys. Then sign up for your open AI account. From here, click on create new secret key. Let's give it a name like n8n Sora 2 tutorial. And let's create this key. Let's copy the key and add it to n8n. Let's save this. And our connection was successful.

[06:39] Now there are two other things you need to do as well. You need to verify your open AI account and you have to add some credit. To verify your account, click on your user icon and click on your profile. Then under organization, click on general. And under the verification section, there should be a button that will ask you to start verification. Now verification is actually a really simple process. So simply go to billing and add some credit to your account. You probably know this already, but Sora 2 is a paid service.

[07:11] And that is why you need to load credit onto your open AI account. And to give you an idea of the pricing, Sora 2, the standard model, will cost you 10 cents per second. But as with everything in the AI space, I won't be surprised if these costs come down in the near future. Either way, with all that said, we're now ready to start using Sora 2 in n8n. Aside from the list of models, we have access to Sora 2 and Sora 2 Pro. Let's simply select Sora 2. Then we can pass in our prompt. And of course, we don't want

[07:42] to hard cut it in this field. Instead, we want to grab the prompt from our form submission and add it to this field. Now we can see the duration of our video. So it could be 4 seconds, 8 seconds or 12 seconds. I'll change this to 8 seconds. Then under size, we can select our resolution, which includes both vertical and wide aspect ratios. I'll just leave it on the default. So let's test this by clicking on execute step. And this took a minute or so to complete,

[08:13] and we can now view this video. And I mean, that's impressive. This video includes music, voices, sound effects, everything. Now we can actually greatly improve these results. Now at the moment, we're passing in a really simple prompt, but there are a few tips and tricks I want to show you to get even better results. The people from OpenAI write this article, this Sora 2 Prompting Guide.

[08:45] So this is all section about before you prompt and API parameters, which we don't really care about. But what we do want is all of this. The section about the prompt anatomy with examples and all of this stuff. So what we can do in our workflow is add an LLM into this workflow that will take our basic prompt and then improve it based on this prompting guide. So what we can do is add a new node in between these nodes. Then let's go to AI.

[09:16] Let's go to OpenAI and let's select message a model. Of course, before we set up our credentials. So we can just go down to the model list and let's select GPT-5 Mini. Now other messages, let's change the role to system. Then on the prompt, let's go to expression and expand this. Then let's say role and I'm just going to use voice to text for this. Your role is to improve the user's prompt based on the prompting guide from OpenAI.

[09:49] The user is attempting to generate a video using the new Sora 2 video generation model from OpenAI. Below you will find an extract from the Sora 2 prompting guide. Please take the prompt from the user and greatly improve it for Sora 2. All right, that looks good to me. Then let's add another section and say Sora 2 prompting guide. Then all I'm going to do is simply copy and paste everything on this page like so.

[10:21] Then let's paste it into this article and then let's add it to our system prompt. Finally, I'm going to add one more section called output and let's say respond only with the improved prompt and nothing else. All right, cool. Now we simply have to pass this prompt into this node. So I'm going to add another message. Then it's going to expression, it's expand this and let's say user prompt and we'll grab the prompt from our forum.

[10:53] Cool. Let's test this. Let's execute this step and hopefully we'll see a much improved version of our initial prompt. And by the way, at this point you can add additional things. So if you only wanted to generate videos that are humorous or want a very specific style like anime, then this would also be a very good place to enforce that behavior. Right. Let's have a look at this prompt. We now have golden hour, low tide beach. And I mean, we can already see that this is way more descriptive. So all we have to do now is go to our

[11:25] Sora 2 node and let's replace this with the text from our LLM and this should give us an even better result. So let's have a look. No, no, come here. Now, of course, the animation might look a little bit floaty and that's just kind of the nature with these videos. You have to do a few generations and then pick the best result. But I can definitely see all these elements coming together, like the golden hour and low tide beach

[11:57] and the details about the dog. Then one final tip I want to give you is that a lot of people share their prompts on X. Because Grok is able to search X for posts, it's a really good way to retrieve popular prompts. And then what you can basically do is copy these prompts and use these as examples in your LLM or as inspiration for your very own prompts. Now, of course, now that we have our video, we can use anything to continue with this automation. So maybe we want to store this video on

[12:27] Google Drive or maybe we want to use this video to build a faceless YouTube channel that runs on autopilot. If you want to learn how to build an automated YouTube channel using N8n and Sora 2, then check out the video over here. Also remember to hit the like button and to subscribe to my channel for more N8n and Sora 2 content. I'll see you in the next one. Thank you for watching.