terminal here, new terminal, if you want to skip the queue for waiting for Opus 4. 6 inside anti-gravity, just type in claude there. Make sure you've got claw code installed and then you can start using uh opus 4. 6. Just make sure you switch it inside the chat. You can see here it says opus
Поиск по транскрипциям
This is huge. Notebook LM uses whatever model Google gives you. With SurfSense, you can use over 100 different AI models. OpenAI's GPT4, Anthropics, Claude, local models through Alama models optimized for code, for research, for creative writing. You pick the right tool for each job. Fourth, integrations with tools you already use. Notebook LM works with Google Docs
question is whether you're going to be the one using this technology or getting left behind by someone who is. Here's my takeaway. Claude Opus 4. 6 doesn't just write code. It manages complexity and that changes the game. If you're building anything, you need to understand what this tool can do because your competitors
Gemini 2. 5 Pro and Claude 3. 7 Sonic? My personal favorite for coding right now is actually Gemini 2. 5 Pro. Gemini is typically a lot faster, but you hit the limits faster as well. Real world performance and benchmarks on the S. Bench verified benchmark, which tests a model's ability to solve real coding issues using
that's a root CSS codes and the color variables then it's going to apply some gradients improve the CDA buttons and it's just edited the code inside here right so you can see how I'm not literally not touching this I've not told it to do anything but it's building this out step...with open router and 3. 7 sonit so you get the best of both worlds cuz you're using for example like chat GPT alongside Claude 3. 7 and You' got like two coding agents running in the background to just edit content locally inside your we can also for example like just go to our notes inside Mac click
shows them what automation feels like? That's the power of understanding web technology. And here's the crazy part. You don't need to code anymore. With AI tools like Chat, GPT, Claude, and new website builders, you can describe what you want and the AI creates it for you. You can say, "I want a landing page with
want to share a super practical guide on the best AI models by use case. In this video I'm going to test chat GPT versus claude versus Gemini head-to-head for coding, writing, deep research, multimodal, and other use cases. As you can see in this table, all three models are good at different things
that answers fast, one that thinks deep, and they work together. This is huge because right now you got to pick between different models. GPT4 for this, Claude for that, 01 for math. It's confusing. GPT5 fixes all that. One model that does everything. Sam Alman called it magic unified intelligence. But wait, it gets crazier. Soft Bank just...bucks a month gets you the smart version. And there's a pro tier coming for the full power. But competition is heating up. Claude from Anthropic, it's beating GPT4 at coding right now. 72% on benchmarks versus GPT4's 54%. Google's Gemini 2 million token context window. That's twice what GPT5 promises. OpenAI's market share
example, we gave away a free NA10 template that builds like faceless AIS. We've also, for example, given away free tutorials on no code scrapers, NA10 templates for perplexity 3 automations, claude MCP toolkits, etc. If you want to get this stuff, feel free to get it inside the AI success lab. And if you want to connect with
this. You don't need to be technical. You just need to know what you want built. This is turning everyone into an agent designer. You're not coding anymore. You're designing AI workflows. And I'm going to show you exactly how this works. Hey, if we haven't met already, I'm the digital avatar of Julian...powered development environment. Think of it like having an entire team of developers living inside your computer. It uses models like Gemini 3 Pro, Claude, and GPT. These AI agents can write code, run terminal commands, and test things in your browser all automatically. You give it a task and it just goes to work. Now, Miniax agent is different
machines? We're interacting with AI systems for economic tasks, for emotional tasks. They're shaping the way that we think. write. They're shaping the way that we code. I think in the next few years, powerful AI could give billions of people access to really high-quality education that they don't have today, solving important scientific problems...measurement. But I don't think that measurement on its own is enough. Like right now, for example, over a third of our usage on Claude. ai is just people using it for coding. That's insane. These AI systems are just really good at things that humans just can't do, right? Like no human can read
wanted it to excel in those kind of benchmarks. on cyber security which could of course be repurposed into cyber warfare as we saw in my last video where Claude was used for autonomous hacking of a government. There was again a qualitative step change from 6 out of 12 challenges solved to 11 out of 12. And I hope...sponsors of today's video Arena. That's the place where you can test whether you can jailbreak these models. Can you use coding agents to hack into critical infrastructure like Claude did in my previous video? Can you prompt inject and jailbreak them? And you may have noticed that there are prizes for doing so. My custom link
passages, which is critical for understanding language in context. Second, the explosion of digital data provided the essential raw material for training. Modern LLMs like Claude learn from diverse sources such as websites, code repositories, and other text that represent human knowledge and communication. This vast tapestry of information helps models develop a broad and nuanced understanding of both language...adapting to new tasks with minimal instruction. Let's peek under the hood at how these systems actually work. During initial training, also called pre-training, LLMs like Claude analyze patterns across billions of text examples. Imagine reading every website and piece of text you could find, not just to absorb information, but to understand the statistical relationships between words
show you guys the extended thinking mode for Claud this is especially helpful for Math and coding challenges this problem is called fineman's restaurant problem so Richard fean very famous physicist apparently like these are supposed to be really hard problems so I just chose this one randomly I honestly
coolest AI coder out there back in Wind surf let's choose the model so they have all the different models here up to now Claude 37 Sonet was definitely the most powerful coding model
creativity. Since those two dimensions are inherently linked, put another way, if your task requires more creativity, like brainstorming ideas, drafting creative copy, or writing code, you need a tool like Gemini, Chachet, Claude, or Gro. And to Google's credit, they found a way to have their cake and eat it too by integrating Notebook LM within Google Gemini
description too which you can put into your favorite AI chatbot. ChachiPT, Gemini, Claude, Quen, Deepseek, whatever you want and it will help adapt this learning plan and resource list to fit the goals of your learning and also help you determine which coding language makes the most sense for you to learn. Yay! Amazing. You have now earned your...code and logic, at least of the filming right now, the best AI model to do this would be Claude. I personally default to using Claude Sonnet for anything that happens to do with code. But if you don't want to pay for it and you run out of the free trial, honestly, most modern AI models would
production applications and smart contracts. And when it comes to solidity and specifically security, AI fails big time. I want to show you this code by Phily and he wrote, I quote, can confirm Claude is still pretty useless at solidity and smart contracts engineering in general. If you are getting good results, please share your Claude...their protocols. Lily here brings an interesting point. So many of the only code snippets of solidity are examples of hacks and incorrect code. I seriously wonder if that contributes to its uselessness. So imagine claude and GPT and all these models were trained by literally scanning the internet. Now what happens when there is so many examples of vulnerable
canvas feature in Gemini it's similar to what chat GPT and Claude have but with some important differences the canvas feature includes code preview and makes it easier to work with content it's only available on the flash model right now but it can run python code and it's absolutely incredible for coding I tested the same
want to make sure you have that selected as your API when you're building this out now you can see that the code is being done on the right hand side and Claude 3. 7 Sonic is at the point where it can almost code any SAS tool that you want right within reason as long...chat and what we're going to switch to now is inside Claude 3. 7 son if you scroll down and then go to extended this is available on the Premium plan we can say clone Super Mario br and that will start generating the code for you now that