Claude Opus 4. 1 update is insane. Today I'm going to show you something crazy. I just put Claude Opus 4. 1 through five brutal coding tests and the results shocked me. This AI just built a full 3D game in minutes. It created interactive landing pages that look like they cost $10,000. And wait until...build a simple website, but when you ask them to create complex interactive experiences, they usually fall apart. So, I decided to put Claude Opus 4. 1 through what I call the impossible test. Five coding challenges that would make most human developers sweat. And what happened next blew my mind. Let me show you exactly what I discovered. Test
Поиск по транскрипциям
Code. The next is better prompts. And I really like this one. after a crappy fix or a crappy build, just tell Claude Code, "Hey, knowing everything you know now, scrap this and implement the more elegant solution. " This is cool because you can conceptualize what Claude is doing is basically searching over a big volume of potential solutions
like the normal clad versus the reasoning clad you know like how do you think about that yeah that's a good question I think like uh the normal Claud without sort of extended I use a lot of document sort document creation document summarization um a bunch of the things that I do around creating prds um creating evals...value on extended thinking is like coding use cases um and we built a product that we launched as well this week in a research preview called Claud code and that's something that's become like integral to the way that we're working internally and a lot of people have noticed the extended thinking being valuable
want to me. If you join the vibe coding academy link down below, you not only get live boot camps on OpenClaw and full video series on OpenClaw and Claude Code, but you can also DM me literally anything you want. You get a direct DM me, you can say the most heinous, disgusting things ever said to anyone
also what you could probably do here is you could install a bunch of Claude code skills. So you could pull in the repositories of skills other people published on GitHub for example for claw code and then you could probably call those inside the chat here as well. I think haven't tested that out. This is the first
created this amazing document. That is incredible. Anthropic just released Claude Co-work and it might just be the most powerful AI tool of all time. It's like Claude Code except it's an AI employee that works for you 24/7 doing anything you want on your computer. I'm about to show you several use cases that will
logged in to Chrome that has the Claude Chrome extension. You have to lo be logged into it with the same account that you're logged into Claude code. So, that's one thing to keep in mind. And now it's asking if we're allowing it to use photop. com. We're going...there's this golden outline around my entire screen. Let me show you. So you can see in the right it pulled in itself I guess the cloud coded desktop it pulled it into the same view as it's working in the tab making the thumbnail. I got to say it's working well now instead of actually clicking
over here called Codex CLI just fixed cloud code that you can watch using the link down below. And basically what I was doing in this video is getting Claude Code to make a bunch of changes and edits to a codebase. And then I got Codex CLI to review those changes. And I think that process is now even
would link it to whatever you want to use. The best thing that you can use for this from what I've seen is the GitHub Clawbot installed with Claude code. Now, it can also run inside your WhatsApp. Here's an example as well. So, you can actually add like Claude, for example, into a group chat as well
would link it to whatever you want to use. The best thing that you can use for this from what I've seen is the GitHub Clawbot installed with Claude code. Now, it can also run inside your WhatsApp. Here's an example as well. So you can actually add like Claude for example into a group chat as well...could have it chatting to you inside Telegram or inside WhatsApp or whatever. But one of the biggest differences here that most people don't realize is like Claude can basically build extra workflows on top of this. So, you don't need to keep going into something like NA10 to build your workflows, right? So I had a friend
would link it to whatever you want to use. The best thing that you can use for this from what I've seen is the GitHub Clawbot installed with Claude code. Now, it can also run inside your WhatsApp. Here's an example as well. So you can actually add like Claude for example into a group chat as well...could have it chatting to you inside Telegram or inside WhatsApp or whatever. But one of the biggest differences here that most people don't realize is like Claude can basically build extra workflows on top of this. So, you don't need to keep going into something like NA10 to build your workflows, right? So I had a friend
would link it to whatever you want to use. The best thing that you can use for this from what I've seen is the GitHub Clawbot installed with Claude code. Now, it can also run inside your WhatsApp. Here's an example as well. So, you can actually add like Claude, for example, into a group chat as well...could have it chatting to you inside Telegram or inside WhatsApp or whatever. But one of the biggest differences here that most people don't realize is like Claude can basically build extra workflows on top of this. So, you don't need to keep going into something like NA10 to build your workflows, right? So I had a friend
from YouTube videos and I can ask it anything within that domain knowledge that I provided. I like that idea. I like the idea of an agent that has Claude code installed and I can say here's an idea for an app and the agent goes and troubleshoots with clawed code until the code that I want is written
that it just writes, it feels nicer. I don't know something about the way it formats text. It's much nicer inside Claude. Bear in mind that open air did recently trigger a code red because they're worried about their competition, Google taking over, etc. And I would agree. I even see that from the YouTube video numbers...specific strategies and it actually remembers to insert a question mark as well. It just feels better. This is why I'm switching over to Claude for pretty much everything or to Gemini for coding cuz GPT 5. 2 and GPT 5. 1 were some of the worst models that I've tested so far and see like
these are like the real developers coming in here. The these are here right now. Uh maybe take a screenshot of that UIUX and share it with Claude Code. Well, we got it working now, so we're good. Beads in Gas Town is a thing, though. Famous Stevie stuff floats on Twitter and blogs, but he's definitely...What do you want to tackle next? Some options. Wire up the assign task button to actually call Claude. Add the codebase scan to visualize files as a terrain. Polish. I feel like we should be able wire up the code base so we can then actually see code changes when we eventually give agents tasks. Do you agree
pretty nice, too. So, this is the one from Gemini. As you can see, this is the landing page. Looks pretty nice. And then this is the page from Claude Code. The only thing that I would say with Gemini is it went a bit overboard on the copy. I didn't really like the copy that's used...sometimes days. And I had to write code, fix it, text it, test it, fix bugs, write more code. It's slow and hard. All right. Um, but now I can and uh build complete SEO tools in minutes using Claude Opus and you know you can use Gemini together with it. All right. So here's what's changed
from Anthropic as of launched fast mode for Claude Opus 4. 6. This is a brand new experimental version just released on February the 7th first for developers via claude code and a limited
started? Well, to get started, I'd recommend um just tell Claude code to get you started. So, you know, I went ahead and I basically was like uh I gave um Cloud Code the list this um this GitHub repo, the GitHub, the auto research GitHub repo, and wow, 25,000 stars already. So, this is crazy
seems that time is slowly coming to an end because that can lead to overusing tools in a certain way and just lead to worst output overall, especially when coding. So bad prompts would be something like you must use a search tool whenever the user asks any question, always search first before responding, never skip searching. That would lead...when you need current information or facts that you're uncertain about. And one thing that the Anthropic team did make and will be linked down below is a Claude Code migration plugin. So basically by installing this plugin, it changes many of the system prompts that may be in your code base to be less aggressive because they
overall. And on web development specifically, it scored 91. 5%. That's better than Claude Sonic 4. 5 and Gemini 3 Pro. This open- source model just beat the top commercial models. Here's why this matters. First, multi- language coding. This thing excels at Rust, Java, Golang, C++, Cotlin, JavaScript, TypeScript, all the major programming languages. Second...handle native Android and iOS app code. Third, is way more efficient with concise outputs and tighter reasoning. And fourth, agent integration. M2. 1 works seamlessly with frameworks like Claude Code, Droid, and Klein. It can manage entire tool chains reliably. Now, let me show you what this