OpenClaude DESTROYS Claude Code?

So today we're going to be testing out OpenClaud, which is an alternative to Claude Code. And this is a free open source coding agent that you can get access to. Right now I'm going to be testing out, you can see an example, it's growing really fast and it looks super powerful. So let's get started with this.

The way we can install it is just by copying this command from the GitHub. So we're going to open this up inside Terminal, let's plug this in. And then once we've done that we'll be able to start it with the command OpenClaude. So let's see what we get in a second.

Alright, so that's installed, took a couple of minutes. And then if we type in OpenClaude, that should work. Happy days, look at that. It looks just like Claude Code, which is pretty crazy.

Now one of the biggest advantages to this is like, this has no limits, right? So for example, if you're using Claude Code and the subscription, well the problem with that is like, with Claude Code and the subscription, there's limits to that package, right? Whereas if you want to keep using something like Claude Code, but you have hit your limits, well you can use OpenClaude instead and then you can use this to code as well directly, right? And so you might be wondering, okay, why OpenClaude?

Well, number one, you can use one CLI across cloud APIs and local model backends. Number two, you can save provider profiles inside the app, right? So if you do forward slash provider, you can actually manage which APIs you've added. And if you want to add a new one, you can just go in like this, right, and just add a new one.

You can even use Olama for this. So you can actually use Olama, which is a local running sort of tool where you can download models and plug it in, right? So if we open up Olama here, for example, you can see we could run like GLM 5.1 Cloud directly with OpenClaude. And GLM 5.1 Cloud is free as long as you stick within the token limits.

And the same, for example, if you're using, let's have a look, OpenRouter. With OpenRouter, you could use OpenClaude as an agent harness and then plug in a free API, for example, like Elephant Alpha. Let me show you an example of what this means, right? So if we go to OpenRouter over here, we're going to select the provider name as OpenRouter.

We've got the API, and then we need to choose our model name, right? Now, if we go over to OpenRouter like so, there's free APIs like, for example, Elephant, right? And we could just use this free model, plug that in, and then just add in the API key. So if we grab an API key from our settings over here, we'll go to settings, manage, open this up.

There we go. And then we're going to select API keys. We'll create a new API key just for this example, copy that, go back to OpenClaude inside our terminal, paste that in. There we go.

And now we're done, right? And so now we can choose our model, right? So we can select our model, which we're going to go with. So for example, OpenRouter Elephant Alpha.

And look at that, ready to work, right? Ready to get started, right? And so now we've got a free API plugged into a free version of OpenClaude. And so we've got an awesome agent harness for coding, and we've got a free API that we have no limits on.

That's pretty amazing, right? And so now we can code whatever we want. So if we go inside here and we say, okay, build out a new landing page for, and then we'll put in the details of my AI Profitable and Community website, just as a little test here. We'll copy that, plug that in, paste it in, hit enter, boom, look at that.

Look how easy this was to set up. This is the first time I've used it, and it's so easy to do, right? And so you get the point, you can get like these free coding tools, very similar to Cloud Code. I'm wondering if this is actually based on the leak from Cloud Code or not, or whether it's been around before then.

Yes, you can see here, it really just got set up early April, right? Friday the 3rd, and then it must have been leaked and rebuilt directly here. And so now it's beginning to build out. And you can see here that it's actually using the same sort of style that Cloud Code used previously, right?

In the way that it's responding, the way that it's replying, et cetera, right? The difference is, it's so easy to plug in your own models. You can manage different providers. You can use different APIs inside there, and you can code with whatever you want, right?

So you can see here that it's working pretty nicely. And also we've got a free API with this. So let me run you through a quick guide on how to use this, what it is, why you'd use it, et cetera. You see it's building out the HTML now.

So let's break this down, right? Why should you care? Well, most AI coding tools lock you into one service, right? So for example, Cloud Code would lock you into using Cloud Code.

OpenCloud lets you switch between 200 different AI models, which means if one AI is expensive, you can just switch to a cheaper or a free one in this example, right? And if one AI goes down, you can use another. You're never stuck, right? So for example, recently, like a lot of people saying, the performance of Opus 4.6 isn't as good.

So you could just switch over to Elephant Alpha instead or use a new API and test it out. Now, what can OpenCloud actually do? Well, it can write, edit, and fix code files for you. It can do anything that Cloud Code can do, basically, right?

You can see here it's deploying sub-agents and creating a to-do list for the task we're running here. But it can write and edit code files for you. It can search the web to find answers. It can run commands on your computer automatically.

It can work with multiple AI agents at the same time, like having a whole team of AI agents. And it streams answers in real time so you can see the results as it happens. It also works on Mac, Linux, Windows, et cetera. So you can see how it's beginning to plan this out and build out the project.

And now it's actually coding this out, as you can see. So I'm going to click on Yes, it'll allow. It literally feels just like Cloud Code, which is pretty crazy. And you might be wondering, okay, well, which AI does it work with?

Pretty much everything, right? Any model out there I think you could use this with. You see how it's ticking off the boxes in the task right here? Why would you use local AI, right?

That might be another question you have. Well, Alarma lets the AI run on your computer, which means that this is free and the data never leaves your machine. So there's two big advantages to using Alarma or local models like that. Number one, you don't pay anything.

Number two, you keep your data private, right? And it works without internet connections. So you could actually use something like Open Cloud to code locally offline, right? You could be on an airplane and still code, which is pretty crazy.

Now, how does it all work? So you install Open Cloud once using a simple command. You tell it which AI to use using the provider section. You can type in your own and then you just type in plain English what you want to build, like you've seen, for example, here.

It's then going to tick off the tasks, as you can see, and you get results live in your terminal, right? It's just one CLI that connects every major AI under one roof. And it's nearly finished on this particular task as an example of that. So to get started with this, you need to install Node.js, right?

Make sure you've got that downloaded. Then you install Open Cloud using this command right here. And I'll put a full guide inside the AI Profitable Room. From there, you choose your AI provider, right?

And the way that you can do that is you just do forward slash provider like this. By the way, it's finished the task now, so let's say it opens up. Or you can use Olama for free as well. You can launch Open Cloud, which you've seen today, and it actually works.

And you can start giving it tasks. For example, like code this, create that, et cetera, right? So we've got the file ready to preview here. This is a website.

And boom, look at that. That's a website created through Open Cloud. It looks pretty nice. We coded it out for free with Elephant Alpha.

It was incredibly easy to set up, right? That was very, very easy, and it was coding fast, right? Because we did that whilst I was talking live to you right now. And then just start giving it tasks, you know, test out, fix a bug, search this, do this, et cetera.

It is really powerful. I've not seen anyone else talk about this, but it's crazy powerful. So to recap everything you need. Open Cloud is a free open source AI coding system you run in your terminal.

It works for over 200 models, including OpenAI, Gemini, DeepSeek, Olama, and more. You can use it completely free of Olama or Elephant Alpha. And that's a new stealth model available on OpenRouter, running locally on your machine. So Olama runs locally, and then Elephant Alpha is available on OpenRouter, as I've shown you today.

And then you can write code, edit files, create apps, search the web, run commands automatically. You can install it with one command. The fastest start is set free environment variables. You can use Provider inside the app to save and switch between AI profiles.

You can use Agent Routing to send different tasks to different models. You can add Firecrawl for better web search results. And you can use Olama for 100% private offline, zero cost AI on your machine. And that's basically it.

You don't need to be a genius to use this. You can just get started straight away. And this is exactly how it works. One thing to say is Open Cloud is not affiliated with Anthropic, right?

It's just like a free open source project that's just been announced and released on GitHub. So thanks for watching. If you want to get my full guide on exactly how to use that, you can get that inside the aiprofitboarding.com. Plus we have tons of community members using Cloud and Cloud Code and all sorts of crazy AI coding tools, Hermes, Open Cloud, etc.

inside here. So if I add a new guide here, you can get the full guide. It's available on Open Cloud. This is my AI automation community where you can get help, you can get support in real time, you can ask questions.

We have a community where you can post questions inside the community feed, as you can see, and personally connect with me too. You can also go inside the calendar and jump on weekly coaching calls where you can get help and support in real time, share your screen, ask questions about your setup. Inside the map, you can connect with people in your local city. Building out AI agents, you can connect with local Cloud users in your local city as well.

And then also inside the classroom, you can get access to all my best training. So feel free to get that link in the comments description or just go to the aiprofitboarding.com. Thanks for watching.