Just stumbled across a video talking about Claude being free, so I had to dive…

You can now run Claude code for completely free with almost unlimited usage. So somebody on github just built a free tool called OmniRoute that links your Claude code to over 200 other free AI providers giving up to 1.6 Billio- Alright I've heard enough, let's go check it out for ourselves shall we? Alright got the link right here, you can see their title and their mission, 290 providers

Alright got the link right here, you can see their title and their mission, 290 providers and more than 90 providers are free. Let's go ahead and clone the repo on our system, give it a little bit and it should be installed in no time. As you can see it has been installed. Create your env config file, do a little more magic in your terminal and at the end paste

Create your env config file, do a little more magic in your terminal and at the end paste this weird text. After everything is done, go to your browser and type localhost 20128 and you will be directed to OmniRoute. And I won't bore you with all the basic setup. Then go to providers and this is where the real stuff begins.

Then go to providers and this is where the real stuff begins. Because you manually have to sit there and find all the free providers. Not only that, you also have to give them your API key you generated from the terminal which I did leave out for obvious reasons. Then I paste the command to redirect OmniRoute's requests to Gemini or other AI models.

And after all the boring stuff, go to combos, press create combo, give it a name, find a provider and a model, test if it works and now you can run Claude for free. So here if anyone has a hard time understanding why we did what we did, Claude code acts as a terminal based developer tool.

acts as a terminal based developer tool. It knows how to inspect your local project directory, edit code files directly, run terminal commands, execute tests and make git commits. And the megamind behind all this would be Google Gemini. The LLM that processes the context, decides what code to write and dictates which terminal

commands to execute. And if you want more in depth explanations, drop a follow.

And if you want more in depth explanations, drop a follow.