Поиск по транскрипциям

model. All right, so I'll be trying out GPT5 and GPT5 codecs via codec cli and also comparing it against each other and to claude opus 4. 1 via cloud code on these two real world production code bases. So, the first of which is an AI news application that I made that basically helps stay up to date

update that came out last week from Claude Co-work, which is like a full desktop app, and millions of people are using it, that was built by Claude Code in a week, right? So, the power of these AI tools is basically limitless if you understand how to use them and if you have the right ideas

They've also got a new update called agent teams as you can see right here. So, you can orchestrate teams of Claude code sessions. I'm excited to try this out and you can also get it available on the API. You see here that Claude Opus 4. 6 six is available on claw. io claude developer platform

something that answers questions like a company expert, something that sites sources, something that actually helps people instead of just spitting out generic answers. This is what people wanted Claude code for. the ability to turn their knowledge into an intelligent assistant, but they wanted it without being locked into Anthropics terminal. Watch Claude work. It's reading the documents

hits a problem, it usually gives you clean error messages and suggests fixes that actually work. It's like having a senior developer looking over your shoulder. But clawed code, when something breaks in a workflow, it can be a nightmare to debug. The error messages aren't always clear, and because it's dealing with so many external connections...recommend starting simpler projects with codecs. You'll learn faster because you won't get stuck debugging complex integration issues. Save those headaches for when you really need Claude Code's workflow powers. Here's something else that might surprise you about these tools. The speed difference is huge. Codeex generates code fast, like really fast. You ask for something

This new Claude MCP update is pretty impressive. So, what I want to show you now is an example of a new update inside Claude Code. You can also do this inside cursor and everything else. And basically, you can access loads of different tools that I'm about to show you using something called the MCP Docker. And essentially

doing something wrong, but I'm coding for three hours. I'd say it reset. What do you mean by reset, Benam? Do you mean like compaction? Like when claude code compacts? Is that what you mean? Stop using superbase. What are you using instead? Can anyone steal man superbase? Yeah, I've been using it for all my apps

have a tool like Teddy's showing you, it's more copying and pasting the right thing than knowing what to type in manually. And some like Claude code can even just run in your terminal and do it for you. Yes. And you know, what's really funny is like Claude is so good that...frequency of, "Hey, have there been changes? " Um and then let you know when there have been changes done to the API docs. Yes, something I did with the Claude project as I brought in the API docs. And if you've done that before, you could just paste in the link and say, "Tell me what's different from

actually looks like AI Studio is a lot faster for the code. How quick that is coming out. It's blasting it out. So, we got the code back from Claude. Let's take this, copy that, and we'll go over to online preview and we'll see what we got back here. So, this is the example from...bear in mind as well like Claude is a lot more expensive, right? So Claude, if you're using that on open router, the prices are like they're crazy for Opus 4. Whereas for example, AI studio, you can actually code for free if you're just using it within the token limits. And if you're absolutely rinsing

being able to chill while Claude cooks is very fascinating. Uh what was I talking about before? Frankenstein cursor or something like that. Yeah, I don't know. Claude code's better. Who cares? [ __ ] Christ. Just try out. Don't try to help me. No one I don't want help. I don't want anyone's help

current cloud code instance. And I'm going to paste this in. And I'm basically going to say, hey, set up my current workspace as a GitHub code space so I can share it with collaborators. Then I'm going to be able to um invite people by GitHub username. And so, you know, they're going to have...just two things you need to do. you need to go over here and then actually connect to your agent of choice. So in my case I'm doing claude code. So I'm just going to click install. Then what you have to do as well is you have to connect to cloud code. So after you've done

game run up was clawed deep SE and Gro didn't work deep SE didn't really code anything and grock's game was so bad I wouldn't even play it if you paid me round two SEO blog ried so Claude actually came out with the most humanized engaging content with a cheeky Touch of humor deep seat...YouTube script writing chat GPT dominates for AI detectability and game development and Gro one for the card game basically chat chap is the best for games and coding Claude script writing and deep seek last almost every single round Savage now if you want the prompts join the AI profit

coding. It actually looks like AI Studio is a lot faster for code. How quick that is coming out. It's blasting it out. So, we got the code back from Claude. Let's take this, copy that, and we'll go over to online preview and we'll see what we got back here. So, this is the example...from Claude Opus. Honestly, not really that much happening. If we try and shoot right here, nothing really comes up in the scores. It looks pretty cool though, to be fair to it. Let's have a look now at the code from AI Studio. We'll grab that. Here we go. That is looking way better. I think

just say to claude okay go off and implement this in the same way that you give it to a normal assistant but the difference here is claude code can do it in the background instantly whereas a virtual assistant they might take days off weekends off they get sick they get tired they forget what to do or they

their lines of code at the beginning of the year to typing barely any of them at the end of the year. I think of this as the claude code experience but for all forms of knowledge work. I also think that probably continual learning gets sold in a satisfying way that we see the first test deployments of home

that they  explicitly mentioned. Also, Claude Sonnet 4 did   introduce a 1 million token context window. So  some people are wondering when it's going to   be available in Claude Code. And it seems that  it's rolling out slowly. So for example, this   user now sees it in their slash model. So if they  run slash model, they...different distractors in your context window   that can reduce the model's performance.   And learning more about what I just said,   then you can watch my previous video about  Claude's new context window. And that should   be linked down below. But if you are really keen  to try the new context window, then Cursor does   support it with Claude

notice some performance improvements when Claude  is searching through the codebase to find relevant   things. And you can opt out of that behavior by  using this command before running Claude Code or   putting that in your environment variables. They  also made it so that @-mention supports files

said now try again. You should see a bright red square where you spawn an agent. Nope. Let's reset the server. Spawn an agent. Boom. Nope. Copy. Claude code. No red square. Here are the logs. Isn't this nice? I got the server with the logs over on the right. So, I can easily copy and paste that...people trolling me? Am I being trolled? I'm just trying to build code. C code commander. People are trolling me. All that talk and you still did not implement it to this game. Implement what? What are you talking about? Give Claude this repo and tell it to use it in the game. You don't need to Google

tokens of padding code, which is just,   I guess, like random related code. And it seems  that Sonnet was faster by about three seconds,   but it did score lower on their own vibe check.   And then they got Claude to play AI Diplomacy,   which is their own variation of the strategy  game Diplomacy. And they say that Claude did   surprisingly...mention earlier in  the video that it's more expensive than Gemini,   so you have to bear that in mind. Now, once it's  released as part of the Claude Code subscription

going over the last week of Claude Code  updates and how you can be using them in your   workflows. First of all, we have Output Styles,  and that basically allows you to adjust Claude