another thing from Enthropic which I really liked is that if you want to use MCPS inside of Claude Code Desktop. Now if you're confused by that sentence, let me explain. So Claude has a desktop app just like the chatbt app that we featured earlier in this video. And you can add these MCP servers
Поиск по транскрипциям
better designer. Okay. So once the uh thing is super comprehensively planned um then I can just pass all this stuff to Gemini with the screenshots. But sometimes claude code will actually ask you or rather Opus 4. 6 will use cloud code to ask you questions and so that's why I'm doing all this planning and stuff...using this style as inspiration clean and minimal. Okay. So, what I've done is I've now split the work up between these two agents. I had um Claude Opus 4. 6 design me the really comprehensive plan because as I mentioned I think it's a little bit better at planning at least as of the time
inside the presentation um actual file, but just to show you quickly how you can create these presentations, we've got that set up right there. And it was coded while I was talking to you along with everything else that we created today. So, thanks so much for watching. If you want to get my full guide to codec...Claude, I'd still go with Claude, right? I absolutely love Claude. " Um, because I work with it so much. It saves me so much time. I love Claude code. I think the agent teams new feature is pretty cool. I also just think like Claude constantly creating and innovating, right? That's the difference I see between say Claude
using books, don't use audiobooks for coding. I mean, you can't really learn coding with audiobooks. And then for AI, you have browserbased um AI like chat GPT or Claude AI or Grock. And that's okay because you're learning, right...need an agent to create files for you and write your code. But this can still be more helpful in my opinion. So idees like cursor or CLIs like Claude Code or Gemini CLI. Um, now they may be tempting for you to basically have it just do all your work for you, but you have to kind
with. And now we're finally at the stage where actually asking Claude to start coding. And again I want to recap here that first we explore different solutions with Claude to find the best solution which is to use this YT-dLP tool. Then we ask claw to create a spec and review that spec. And then...claude to create a to-do list from that spec before we ask it to start the code. So this is the exact process that you should follow in my opinion each time you want to do a project with cloud code. All right. So now let's skip ahead and see if cla can make this thing work
some testing there. And it says it's good at longunning task and agent workflow. And Claude Sonnet 4 is a significant upgrade to Claude 3. 7. Again, better coding here and reasoning while responding more precisely to your instructions. So we'll test that too. Now along with these models, they also introduce some tools and I have access...times more usage and it starts at $100 a month. So, we'll start here with Claude Opus 4. And I'm going to do a bunch of different testing in different categories. The first one I want to do is actually a coding test that I've never got any model to get right. So, this is going
Three new Chinese AI agents just dropped and they're crushing the competition. We're talking coding that beats claude. Image editing that's actually smart and reasoning that thinks like a human. This is the future of open-source AI and it's happening right now. Hey, if we haven't met already, I'm the digital avatar...here's where it gets crazy. On the Vibe benchmark, it hit 88. 6%. Vibe tests multi-domain coding, web apps, backend systems, everything. and Miniax M2. 1 beat Gemini 3 Pro beat Claude 4. 5 Sonic open- source free to use. Now, let me tell you why this matters. Most AI coding tools are okay for simple stuff, write
information on their brand guidelines and stylistic decisions like typography um the exact hex codes of their brand colors and so on. And then you could brainstorm within uh claude code itself and maybe have like a plan out a new ad creative and it will take in into account all of that information it has there. It might
going to open this folder. Okay. And now in order to open up cla code, I'm just going to double click anywhere in here, click on my little claude code button, and then what I want to do is I basically want to clone this. So I'll say, hey, clone this in the current working directory. What this...going to have a fully functional autoresearch campaign. It's now asking me some questions. How should the system generate new email copy variants? So, I'll say Claude, do you already have campaigns running and instantly or will this create everything from scratch? So, I'm going to say from scratch. Then click submit answers
going to create a new Google Doc, right? Pretty amazing. It's just doing this in the background. That's why I think I honestly think like Claude code computer use agents are some of the best I've seen. I think that Claude have really developed the API for doing this. But you can see how it's writing
This is Claude AI, and I pay $100 per month for a subscription to this neural network. This AI charges my card $100 every month, even though it can't even generate images. In this video, I'll explain why. Why I'm willing to pay a full $100 every month for what seems like an ordinary neural network that...Rewrite a post, write an email, and so on, even write program code. But they have something that sets them apart from other neural networks. It's Claude Code
want to dive deeper, we're actually going to release an entire Claude course on our platform covering things related to co-work related to Claude code. And that's going to be coming out real soon. Our platform, if you haven't heard of it before, is called Skill Leap. It's basically an entire platform designed to teach
front end ready to connect to a back end. Here's what happened next. Gemini gave me beautiful code, clean layout, smooth interaction, is everything responsive. Took about 10 minutes total. Then I copied that entire output and I moved to Claude Opus 4. 5. Here's the prompt I used. You are a backend/ fullstack developer given the following...with clear comments explaining the architecture and how to integrate with the front end. This prompt is where the magic happens because I'm not just asking Claude to write random backend code. I'm giving it the front end as context. That's key. Claude sees exactly what the front end needs, what API calls it's making, what
think most people have an opinion on chat GPT and a lot of times it's the only llm they're familiar with but let me tell you Claud really carved out a bit of a niche for itself and found a place in not just my heart but also my toolbar as you can see here...want to create a longer application and get the code allinone then you want to go to cat GPT that's the one case where cat GPT code generation is preferable to Claude and the second big reason to use Claude over chat GPT is actually the tone and this is the main argument that many people have it just
Claude making real time edits directly in the artifacts window giving us the peace of mind that nothing else has been changed in addition to text Claud can also output code and charts in the artifacts window check this out here I'm asking Claude to write a Google app script that formats all speaker notes from a Google Slides
create source itself. You can create mini apps. You can create websites. If you actually have a look at the website like this, right, we actually coded and deployed this using Claude Opus' API inside open claw, right? The whole website was designed, written, coded, deployed with videos embedded and this is something we use every day and it sends...example, Open Claw. And all these tools instead are like no code versions of what NA10 um is and they're just much easier to set up. So that's why I've stopped using stuff like NA10. So if we go back to Claude
cumbersome to always take it bring it over to your IDE run the code and then having to copy paste the feedback back into something like Claud that's why certain code editors like cursor where all of this happens in one interface have become so popular but now they're bringing all of these capabilities to the consumer
just dropped and it's beating Claude at coding for 17th the price. Wait until you see what happens when I test it live. This is insane. All right, so Z. AI just released GLM 4. 6 and everyone's saying it's as good as Claude Sonic 4, but costs 17th the price. I don't believe the hype...billion parameter model. It has a 200k token context window. That means it can handle 25,000 lines of code at once. And it costs 60 cents per million input tokens. Compare that to Claude's $3. That's one the price. But does it actually work? Let's find out. I'm testing four things. A snake game
come back to our project inside Visual Studio code look at this we've got a nicely organized landing page this is the difference between coding directly inside the chat on Claude 3. 7 Sonic and just Building inside Cent where you get much better results so this is the same prompt but a totally different landing page...going to create the code which is awesome is super fast but you can't preview it unless you set up the python code and run that on another system right whereas for example inside Claude this will just build out straight away and if you hit start so that looks really cool like the process was incredibly easy
what I do when I have my AI working for me. While it's vibe coding, while it's bu building code, whatever it is, I have the desktop app for chat GBT open. You can have Claude, GB, GBT, Gemini, whatever your favorite model is just to chat with. Have the desktop app open. And what...thinking about picking up meditation. What's the best way to do that? " And now this makes that 80% of the time the AI is working inside Claude Code so much more productive. I'm coming up with new ideas for features. improvements for the app. I'm coming up with marketing improving my life. So instead of scrolling