Open NotebookLM: FREE Local NotebookLM AI Agent!

So today we're going to be running through Open Notebook LM which is basically like a local version of Notebook LM you can run on your laptop. So like a competitor but the difference is you can actually use it for free running it locally and it's fairly easy to set up. Now the way that I'm going to set this up is I'm going to run it directly inside Claude code. So if we take the hub from this as you can see and you can see the comparison here. Open notebook versus Google notebook. So open notebook selfhat hosted. So you've got your own data. You can use 16 providers including free options from OAMA. You can actually choose between four one to four different speakers whereas with Google notebookm you can only pick two speakers. You get context controls you get three granular levels whereas with notebookm you don't get that. Also with content transformations right you got customer builtin whereas with Google notebookm you actually have limited options. Also you can choose your own API whereas for example with notebookm you can only choose Google models as well. Right. And then also you can customize it a lot more and you can deploy it with Docker as well. So let's

get straight into this. The way that I'm going to run it is just through cloud code which tends to be the easiest. So I'm going to go into the terminal here. I've opened up clawed code as you can see. We're going to continue and then I'm going to say okay run this. We're going to start running open notebook length. The reason I do this is like I just I hate messing around with terminal instructions and so like claw codes tends to open it up a lot quicker. And then you can see it gives me a bunch of options. So, I'm going to hit yes, let's go. And I'm just going to open up Docker as well in the meantime, too. If you

want to use Docker, you can actually install it via docker. com. And it's free to download, as you can see. Literally takes 2 minutes Then install. Then we can go back to the terminal. We can choose which AI provider we want to use. So, I'm actually going to open up Olama, which is free and local. And you can see like the cool thing I like about using code to do this is it's just so much easier and faster, right? It gives you the options. For example, I can select between AI or Lama. Olama is a free model you can run locally as you can see and if you don't want to use claw code or you don't use claw code then just copy and paste the GitHub into claude and you can go from there. So we're

going to continue that's going to set up the docker container as you can see and it's actually set up and installed docker as well. The other thing I like about this method is I don't have to set up everything manually. So it gives me the API options and then I just pick the one that I want to go with and then if we go on to docker you can see that claw code is set up open notebook and then if we open it up from docker you can see we have open notebook set up pretty easy right that is so much easier than the old days when you used to run uh terminal and figure it all out yourself. So now we're going to click yes proceed and if we go to the model section you can see we just need to configure this and then it actually gives us the instructions inside claw code to set up a lama. So make sure you've got alarm set up as you can see. And then we're just going to start adding it as a provider. So we go to settings and if you don't want to configure it directly, you can just say to claw code, can you configure OAMA to run a notebook? As you can see now, it's going to pull in the model and it's running with Gemma 34B, which is a local free model, one of the fastest. Burn says, I didn't know you could do that with claw code. Yeah, it's amazing. Gamechanging. I only found out recently as well. And you can see it's pulling in another model as well. So it's pulling in 57 into llama and then what it will do once it's finished is configure open book to use. Now if we go to open notebook you can see o lama is configured right there. Beautiful. All right. One of the things that really helped here was if you go to your models here inside open notebook click on learn how to configure copy all of that and then paste it into claw code as you can see right here to configure it with lama. it will shut down the docker then reset it up with a llama so that all the models are configured and then we can just add the model. So select a provider and we can use Quen 2. 57B or Gemma 3B for this. All right, so let's try this out. I think Quen will be a little bit slow whilst I'm live streaming. So I'm just going to go with Gemma for now. So let's try that. Put it in the model name. Then we'll add an embedding tool as well. So for embedding we can use XY. Let's try that. I'm just going to open up Alama and make sure we have XY set up. And if you want to just make sure this is set up correctly, you can just go into the ask and search section here. If you type in test, are you working? Hit ask. And it should be using Gemma 34B to set this up. As you can see, right, so you can see the user is asking if I'm operational. There we go. It's actually working. Now, if you want to

use text to speech as well, so open notebook can also generate podcasts, then you can use one of these models. But bear in mind like text to speech would have an API cost. Whereas for example like embedding, transformations, tools and chat, these are all free because you can use other. So it's up to you if you want to use that or not. So now we've got that set up with claw code relatively easy. It does all the technical stuff for you. You can now start using it, right? So you can add sources, you can add notebooks, etc. Anesto says, "What software do you use to record your videos? " So I use Streamyard. Streamyard is the one that I start using. Coach Chrome says you Google Pro. Okay, cool. So we can add new sources here. If we go to not notebook, let's do a test run. Then we can add sources and write notes, right? We can also chat with the notebook here. So let's add a source. I think you can add an existing source as well. So if you've added sources previously, you can. If we add new source, you can choose between upload text and links, right? So you can actually add website URLs into here. Let's try this out. And then if we go into sources, you can see that it's loading up the source, right? So once you add a new source to a notebook like for example this one, you can see that it's pulling in and it's actually scraping the web with that information. So it can connect to the website. It can grab all the information from that site. As you can see, it's pulled in, for example, some of the images, the text from that website as well. And it connects locally to that page, which is pretty nice. And then if you go inside the notebook here, you can add more information and that sort of thing. Right. So now we've got this notebook that's trained up on me, my website, etc. What does Julian do? This is a test run. We'll see if this chat can connect to that and speak to the sources. Now, if you want to use a podcast feature, what I recommend using is you could use chat GBT's API, which obviously is paid, or you could use 11 Labs API as well. It's a small cost per podcast, but it's going to give you like the best quality output to be honest. And there you go. So, it's actually taken the information from the source, connected it. It's called me a fascinating, incredibly talented individual. I'll take that. And then it's got some information about me, what I do, my page, etc. So you can take my website there. So if we go into this chat now, okay, create a lead magnet for Julian Gold's website, see the attach source, use that and generate new lead magnet. All right. So it's taking the information from the source there and then we'll get it to create a lead magnet for us. And depending on how powerful your laptop is, that will decide how fast it's going to respond, right? But it's creating the lead magnet as you can see. And there we go. So you can see it's created a lead magnet. The idea it's generated based on my link building websites. I have an SEO link building agency. It said, "Here's a headline. Here's a sub headline. Here's the idea for the lead magnet. " We just fully automated that using the sources that were plugged in. Now, obviously, you could do a lot more with this. You can generate podcast, you can connected to the web, you could add like 10 or 15 different sources to your notebook and then generate something even bigger. But you can see how easy it is to set up, run. Bear in mind like if you're using open notebook LM then you get more priv whereas if you're using notebook LM directly on the website that's attached to your Google account all your data is stored in the cloud etc. Also you can choose free AI providers with unless you can run this locally free and then you get a lot more customization inside open notebook that you wouldn't get inside your average sort of notebook. And then if you go to advanced you can change some more settings. You've got your settings right here as well. You can do more transformations. So like key insights and stuff like that. So you can go to the playground here and run some transformations and that sort of thing which is pretty cool. And then you can pick between all your models and just customize it how you want. So you might be like I want to use notebook but Claude cuz I love the writing from Claude. I think the outputs are much better. Then you can configure Anthropic inside the settings here and just use that instead, right? Or DeepS or 11 Labs for example as well. The other cool thing with 11 Labs is you can choose more speakers, right? So when you generate a podcast inside notebook, typically you're only going to get one to two speakers inside the audio overview here, right? So you're quite limited. If you use open notebook LM, then you can generate more voices, more different voices, more models, etc. So there's just a lot more customization

using this. So thanks so much for watching. What I've actually done is included all the links and the resources from today. I've included a full framework on how to use this a 30-day plan for mastery and also we've included 100 prompts you can test using this process. So if you want to test out you can get all the free resources from today inside the AI success lab. This is a community that connects you with 43,200 people. If you go to the classroom here then you go to January you'll find the trainings on open notebook as well as a full video tutorial on how to use this stuff. Bear in mind like this whole community is better than most people's paid stuff. It connects you to more people. He's got more video tutorials, gets updated every single day. We include all the prompts, frameworks, 30-day plans for pretty much every single tool out there. All right. If you want to connect with me personally, if you want to get the best automations, etc., then you can go inside the AI profit boardroom. And inside this community, we have 2,000 serious AI builders who are using AI to win. You can people sharing the wins right here. You can see that it's a very active community. We have a daily accountability check-in post as well. This is pinned at the top. This was just 32 minutes ago. and people already posting their goals and their checking and their accountability. On top of that, inside the classroom, if you want to learn how to use AI for business automation, you can check out our sixeek masterass. If you want to learn how to automate AI avatar videos, you can check out this section. If you want to get more clients, then check out the agency course here. If you want to get AI SEO automations, we have a full section on that as well as a YouTube AI road map. Top of that, you can get coaching here. So you can jump on live coaching calls, get help and support from AI PhD trained specialist on a live call plus meet the rest of the community. And if you struggle with overwhelm, confusion, too much information, shiny object syndrome, then we actually have a full focus protocol on exactly how to stay focused based on what's working for me. You can get that all inside the AI profit boardroom.