The AI Firehose Podcast
← All episodes
Episode 1 · May 16, 2026 · 08:22

NEW NVIDIA Palm Sized AI Computer is INSANE!

NVIDIA's New Palm-Sized AI Computer is Insane!


NVIDIA just released the Jetson Orin Nano Super, a tiny device that runs full AI models locally without an internet connection. Learn how this 'Raspberry Pi of AI' is revolutionizing robotics, privacy, and the future of edge computing.


00:00 - Intro: NVIDIA's Palm-Sized AI

00:28 - Meet the Jetson Orin Nano Super

01:05 - Insane Specs & Local LLMs

01:58 - Cloud AI vs. Edge AI

03:05 - 5 Mind-Blowing Use Cases

05:05 - Hardware Limits & Challenges

06:35 - How to Prepare for the AI Future

Full transcript

New NVIDIA palm-sized AI computer is insane. NVIDIA just dropped a tiny AI computer that fits in your hand. Runs real AI. With no cloud.

No internet. Nothing. And it's blowing up the internet right now. In this video, I'm going to show you the new NVIDIA palm-sized AI computer that's changing everything.

Stick around, because what this little thing can do is wild. Okay, so let me explain what just happened. NVIDIA, the biggest AI company in the world, just released a new AI computer. But this one is different.

It's tiny. Like, it actually fits in your hand. They call it the Jetson Orin Nano Super. And here's the crazy part.

This little board can run real AI models. Like, the same kind of AI you talk to online, but it runs them right on the device. No cloud. No big servers.

Just this tiny thing on your desk. People are calling it the Raspberry Pi of AI. If you don't know what a Raspberry Pi is, it's a tiny computer that hobbyists love. It made computers super easy for everyone.

Now, NVIDIA is doing the same thing, but for AI. And I think this is one of the biggest AI stories of the year. Let me tell you why. First, the specs.

This thing pumps out 67 tops of AI power. Tops just means trillion operations per second. So basically, it's doing trillions of AI math problems every single second. In a board the size of your palm, it used to do 40 tops.

NVIDIA just pushed a software update. And boom. 1.7 times faster. If you already own one, you just update it.

Your old board becomes a super board overnight. The memory speed is 102 gigabytes per second. It only uses 25 watts of power. That's less than a light bulb.

And it can run a model called Lama 3.1 with 8 billion parameters. That's a real full-size AI model running on a tiny board. In your hand. Imagine that for a second.

Big AI companies use giant data centers. Huge rooms full of GPUs. Cloud fans. Tons of power.

Now you can run AI on something that fits in a lunch box. That's why everyone is freaking out. Here's what most people are missing. This isn't just about the hardware.

It's about a huge shift happening right now in AI. And if you're not paying attention, you're going to miss it. See, for the last few years, all AI ran in the cloud. You typed something into chat GPT.

Went to a server. The server thought about it. Then it sent you back an answer. That worked.

But it had problems. Problem one. Let's see. Type goes to a company server.

They can see it. Train on it. Problem two. Your message has to travel to the server and back.

That takes time. Problem three. No internet, no AI. If your Wi-Fi is down, you're stuck.

Problem four. You're stuck depending on the cloud forever. Hey, if we haven't met yet, I'm the digital avatar of Julian Goldie. He runs an SEO agency called Goldie Agency.

While he's busy helping clients get more leads and customers, I'm here to keep you updated on all the latest AI stuff. Julian reads every single comment on these videos. So drop a comment below. Now think about this new NVIDIA board.

It fixes all four problems. The AI runs right there on the device. Nothing leaves your room. It's instant.

It works offline. And it just keeps working. This is called Edge AI. And it's going to be everywhere soon.

Your fridge, your car, your camera, your phone, your robot vacuum, all running AI locally. That's the shift from cloud AI to Edge AI. And this little NVIDIA board is leading the charge. Now let me show you what people are actually doing with this thing.

Because the use cases are insane. Number one. Robots. Robots need to make decisions fast.

Like super fast. If a robot is walking and there's a wall, it can't wait for the cloud to tell it to stop. It needs to think right now. Local AI is the only way.

This Jetson board is perfect for robots. Number two. Drones. Same thing.

A drone flying through trees can't wait for Wi-Fi. It needs AI on board. Companies are putting these Jetson boards into drones. So the drone can see, think, and fly.

All by itself. Number three. Smart cameras. Imagine a camera that knows the difference between your cat and a stranger.

Without sending video to the cloud. That's local AI doing its job. Number four. AI helpers at home.

People are building their own AI assistants. Like a private chat GPT. That lives in their house. That nobody else can see.

This board can do that. Number five. Factories. Big factories use cameras and AI to check products.

They need it fast. They need it local. Jetson handles it. Now let me get into the part that really matters.

How fast is this thing really? Vidya says it can run Llama 3.1 8b at around 20 to 30 tokens per second. Tokens are like word pieces. 20 to 30 tokens per second means it's typing out AI responses about as fast as a person reads.

Maybe faster. That's wild for a board this small. For comparison, your phone struggles to run small AI models. A laptop can do it but uses a lot of battery.

This Jetson sips power. 5 watts. And it just chugs along. And listen, if you want to actually use AI in your business, not just watch videos about it, come join the AI Profit Boardroom.

That's where I share all my workflows for stuff like this. We have weekly live calls where we cover real use cases. Like how to set up your own local AI assistant for your business. How to use edge AI like the Jetson board to automate tasks without sending your data to the cloud.

How to build AI agents that handle your customer support. How to plug AI into your daily work so you save hours every single day. We have a full community of business owners doing this stuff right now. The link is in the description and pinned in the comments.

Come join us. Now I want to be real with you. Because I don't want to hype something and have you regret it. This board has limits.

It only has 8 gigs of RAM. That means it can run small to medium AI models. Not the giant ones. You're not going to run GPT-4 on this thing.

You're not going to train a huge model on it. It's made for running smaller optimized AI. Also, you need to know what you're doing. This isn't plug and play like your iPhone.

You have to install software. Use the command line. Tinker. If you're brand new to tech, there's a learning curve.

For developers, students, robot builders and AI nerds, this is the dream. Tiny. Powerful. Easy to ship in products.

Easy to play with at home. And here's something most people miss. Vidya isn't just sharing a board. They're sharing a whole ecosystem.

Free software. Free tutorials. Whole community. A place called the Jetson AI Lab where people share projects.

They want this to take over. And it just might. Think about it. Five years ago, AI was something only huge companies could touch.

You needed servers. Engineers. Tons of resources. Today, a regular person can run a real AI model on a board the size of a deck of cards.

That's the AI way of actually reaching normal people. Not just big tech companies. And this is just the start. Imagine what these boards will look like in five years.

They'll be 10 times more powerful. They'll be everywhere. In your glasses. Your car.

Your watch. Your kitchen. Have local AI. Think.

Every device will see and hear and respond. That's the future Jetson Huang, the Nvidia CEO, is building. And honestly, watching it happen in real time is wild. So what should you do with this info?

A few things. One. If you're a builder or developer, look into Jetson. Start small.

Get one. Play with it. Build something dumb and fun. Then build something useful.

The skills you learn now will be huge in two years. Two. If you run a business, start thinking about how local AI could fit your work. Maybe you don't need to send customer data to the cloud anymore.

Maybe you can run AI in your office. On your own device. That's a real edge over your competitors. Alright.

Quick recap before we wrap up. Video dropped a palm-sized AI computer. It runs real AI models. Clearly.

With no cloud. It does 67 tops. This is 25 watts. And it's starting a new wave of edge AI everywhere.

Robots. Drones. Cameras. Home AI.

All of it is going local. This is the Raspberry Pi moment for AI. And we're living through it right now. And listen, if you want to actually use AI in your business, not just watch videos about it, come join the AI Profit Boardroom.

That's where I share all my workflows for stuff like this. We have weekly live calls where we cover real use cases. Like how to set up your own local AI assistant for your business. How to use edge AI like the Jetson board to automate tasks without sending your data to the cloud.

How to build AI agents that handle your customer support. How to plug AI into your daily work so you save hours every single day. We have a full community of business owners doing this stuff right now. The link is in the description and pinned in the comments.

Come join us. And if you want the full process, SOPs, and 100 plus AI use cases like this one, you can also join the AI Success Lab. It's free. Links in the comments and description.

You'll get all the video notes from there. Plus access to our community of 38,000 members who are crushing it with AI. Drop a comment below. Tell me what you'd build with one of these tiny AI computers.

Julian reads every single one. I'll see you in the next video.

More episodes

Browse all episodes →