Open NotebookLM: FREE Local NotebookLM AI Agent!

Today we're going to be looking at Open Notebook LM, which is a free local Notebook LM AI agent. You can basically install and use for free locally on your laptop. So, if you never used this before, I'll be going you through it and showing you exactly how to use this. Today, let me guide you through exactly how it works. You can actually download this for free directly on your laptop and then you can run it locally on your laptop, right? So, I'm going to guide you through exactly how it works, how to use it step by step, and how the whole process works. If you need help installing it, I'll show you exactly how to get started with this as well cuz it runs locally. So, it is a little bit more difficult to use than using something like notebook. But the good news is here you can run open notebook free locally on your laptop and it's quite easy to set up. So, the first

thing you're going to need is Docker, right? You want to make sure you have Docker set up. You get it for free at docker. com. And by the way, if you want all the setup notes from today, feel free to get it inside the AI Tesla. This is a completely free community with 40,000 members that shows you exactly how to use AI automation. If you go inside the December 2025 section, you'll get all the notes on open notebook LM and how it's set up, right? So, what you can do is you can install Docker and then from here you can use this as a powerful open source AI powered note-taking research platform that is great for privacy as well. So, you can host it locally. You can also generate podcasts, AI powered nodes, you get privacy control, you get content integration, etc. You can generate audio, that sort of thing. Pretty crazy, right? You can install it locally. So, let's get on to exactly how to use it.

So, the first thing I'm going to do once you've installed it is you're going to open up Docker. So, make sure you have Docker set up as you can see. And then also the other thing that you want to make sure you have is Lama. Alarm again is free as well. So, you can download Alama. And then what you're going to do is go to models and make sure I'm going to be using Gemma 3. Gemma 3 is the model that I'll be using today. From there, you want to make sure you installed Alama Gemma 3 1B. Make sure it's running. All right. So, let me show you exactly how to set this up. So, you go into terminal and just pull in Lama and MX Spy by as you can see, right? These are the models that will be running locally on your laptop as you start using this stuff. From there, you want to open up terminal again and then plug these instructions in, right? And then we're going to run this. All right. And then we should be able to access this via here. If we can't, let me go back to Docker over here. Click on play. And there we go. We've got it all up and running. Right. So this is open notebook running locally on our laptop and it's running for free inside OAM. All right. So we've got MXY selected as the embedded model and we've got Gemma 3. 1 running as a tool model. By the way, if you haven't used O Lama in general, pretty powerful. Let me show you this. So this is an example of Gemma 31B running on my laptop as you can see. And it's going to say are you working? You'll see how fast this responds. So you can see how quick it is to reply. It's pretty awesome. If you're fine or something like that, it's pretty cool if you're just running this offline

locally. But we're going to have Gemma 31 working in the background. And then I'm actually going to delete these models just to show you how they're set up. Right? So I'm going to delete that and delete that. Right? Just because I want to show you how to set these bad boys up, right? So we're going to add a model. You're going to select a provider which is a lama. How do you get a llama working with that? You follow the instructions inside the video notes from today inside the AI success lab as you can see. And then if you go inside here, we're going to take this information. And now we're just going to plug in the model name. Right now all you need to do, it's pretty simple. Just plug in the model name like that. Add model, right? And that's going to add the language model. Then we're embedding model, right? Which allows you to do semantic search. So how do you do that? You're going to use mxy embed large. So we're going to add model there. I'm running this locally on a MacBook M3 just so you know. But yeah, you can see us running that now. Now what we're going to do is select gemma 3 for the chat model and the tools model. It's important that you set these otherwise this will not work. All right? So make sure that you do actually use this stuff. All right. So if we go here now, we've got MX buy set up as the embedded model, Gemma 31 as the transformation, chat model, and tools model. Then we're going to go on ask a search and just test this out. And you can see here, for example, inside the ask section, I've run a question and it's given us an answer down there. All right. So if I say just to check this is actually working and connected. If I say okay, what was the score between Man United and Aston Villa? We'll click on ask and we'll just check if it's actually working. You see, it's given us

the results. All right. So, we're going to go to notebook limb over here. And then what we're going to do over here is we can plug in our notes and our sources, right? You can actually generate podcasts as well as you can see. So, you can actually generate podcasts which is pretty crazy. And if we go to notebook here, we can run this, right? And so, what I can do is create a new notebook, enter the notebook name. So, let's say for example, let's load in sales SP for training my team on sales, right? We plug that in. And then over here, we're going to add a new source. Add a new source over here. And we can upload a document, right? So, if I go to my sales training SAP, grab that as a word document. We'll download it and we'll upload it now as a file. All right. So, we'll upload that file. That's beginning to load. We're good to go on this. All right. So, now what we're going to do is click done. And that's going to import this as a source. Now, it will take a little bit of time to import that, right? But you see that's working in the background over here. All right. So, we've imported the sales SP now and that's processing as a source. All right. We can actually collect all our sources over here. So, you can see for example, I've actually loaded in the website as well. And the way that I did that is I just uploaded my page. So this my actual website and then I actually created a whole page right here. And again like this is very similar to notebookm but it's the open source version right and it runs locally and it's more private. That's the

difference. Let's talk about the old way versus the new way. Right? The old way was like your documents were spread across Google Drive notion and random folders. You were searching through hundreds of files to want find one piece of information. You're uploading potentially some people uploading like sensitive data to third party AI tools hitting usage lift right when you need the tool most paying fees for like monthly subscriptions researching on someone else's server or starting from scratch every time you work on a new project with open notebook LM right you got one place for your documents and research AI remembers everything that you've uploaded your data never leaves your computer there's no limits on how much you can use it is free and open source it works offline and you own and control everything right so that's the difference that's why this is different right here All right. And you can see how it works and it's available on top as well. So it is open source. If you have any questions about this as we go along, feel free to ask. But inside this notebook, we can also add the source. So I'm going to add an existing source. We'll take the information from my website. We just imported over here as selected. And then now we can chat with the notebook, right? And so you can see here we can select our model. I'm going to use Gemma 31 which is the fastest and the easiest. We say okay tell me the 8020 about this document. We'll hit enter. And then from here we can start using this. All right. We can also ask and search. So we can actually ask our knowledge base information based on our knowledge base or we can search across our knowledge base for specific keywords or concepts right and we can search for text or vectors as well. Top of that we can choose podcast. So we can actually generate podcasts as you can see and you can see that's generated the podcast right here with different personalities. So we've got the voice ids and the person etc over here. And then inside the model section you can actually use loads of different models. So you could use for example, you could use Gemini. There's actually a bunch of others. So you could use anthropic 11 labs as well for generating audio. So you could actually generate podcast with your voice using 11 labs and the voice ids inside there. We could use deepseek. We can change and mix around with our models right here. So we can change our text to speech model using the data from 11 labs. And yeah, you can see how easy this is, right? You can also do transformations. So for example, key insights, dense summary, analyze papers, reflections, simple summaries, etc. You've got the playground over here. And inside here, you just get a bunch of settings that you can change. But again, amazing tool. You can use it locally. It's completely free. You can use free APIs to work with it. So, for example, Lama like we use today. And then you can just build with this stuff, right? And it literally takes a few minutes to set up and start using. So, coming back to

this, I would use something called the Goldie local AI framework. I'm just going to close this so it doesn't slow down my laptop completely. First, you can install OAM and pull the models. That takes about 10 minutes. One time setup. Then, you're going to organize your knowledge. So think about what knowledge will be most valuable to have instant access to your best content, your SBS, your research. Then from there you can connect everything. So run the docker command and connect to the interface. And this really where the magic happens. Then you can add your sources. So for example, you can upload documents, PDFs, text files, your notes and build your personal knowledge base. And then finally, you can learn and iterate, right? So you can ask questions. You can see what works. You can refine your system over time. And if you want a 30-day implementation plan that walks you through exactly how to master this in just 30 days, then feel free to get that inside the AI success lab and you've seen exactly what we can do and how we can do this. So, homework, watch the video replay from this, install a llama on your computer, pull the Gemma 3 model in, make a list of the top 10 documents you'd want in your knowledge base, and post your progress inside the AR profit board community. All right, link in the comments description to this and inside the community here. You can post your progress, share your wins, ask questions, get help, get support, get advice. There's actually unlimited tech support in there, so you get help whenever you need to. You can also jump on the coaching course if you ever get stuck. And inside the classroom, you get access to my best trainings, SAPs, templates, workflows, etc. Now, if you sign up, you also get all of this stuff as well. So, you'll get, for example, my AI avatar system to clone yourself and scale content without being on camera. You've got, for example, my Twitter automation systems that show you how to reach millions of people per month using Twitter. based on what works for me, my Empire Builder Focus manual, free SEO boardroom access, the digital nomad blueprint to how I built a business whilst traveling the world and also my systems for building school communities as well. So, you can get that all inside the profit boardroom along with my pre-built client funnels, AI outreach engines, YouTube faceless avatar systems, done for you client delivery workflows, ways to get clients, right, trainings on how to get clients. So, feel free to get that link in the comments and the description to the AR profit board. Hope to see you inside there. Plus, you can DM me and get help whenever you need