How To Install Fabric - Open-Source AI Framework That Can Automate Your Life

fabric is an amazing open-source project that promises to give you all the tooling you need to solve everyday problems using artificial intelligence and you can kind of think of it as a library of tried and true prompts that work incredibly well these are Community generated prompts and reviewed that really work on a wide array of different problems that you may face so I'm going to show you how to install fabric then I'm going to show you how to use it so quickly this is fabric I'm going to drop the GitHub URL in the description below so you can find it yourself it's completely open source completely free you can use it with gp4 you can use it with any open source model you can run it locally anything you want and it covers a lot of different use cases that are being added to and iterated on all the time so here's just a few examples extracting the most interesting parts of YouTube videos and podcasts writing an essay in your own voice summarizing opaque academic papers creating perfectly matched AI art prompts for a piece of writing explaining code Cod turning bad documentation into usable documentation and all of these are done with incredibly detailed prompts that work well so all you need to do is first open up your terminal and we're going to get clone the repo so I'm on my desktop I just type get clone and then the repo URL so I'm going to hit enter and it's cloning the repo to my desktop now we're going to CD into it so CD fabric now you need poetry for this and I actually struggled a lot getting poetry working on my machine but I think that's because I had a really kind of messed up python environment if you have any issues drop your problem in the comments below hopefully I'll help you you can also jump into my Discord I'll drop a link to the Discord in the description below now if you don't already have poetry installed this is how you install it so curl dssl The Poetry URL right here pipe Python 3 Dash so I'll go ahead and do that again but I already have it installed yep so already installed but if you don't it'll install it for you now the next thing you have to do is just run the setup script so dos setup.sh and then hit enter all right it's done so we see this command TS not found and I think you can just ignore that I haven't found any problems by having that issue so to apply the changes please run the following commands in your terminal this will just restart your shell but you could also just simply open up a new tab I'm using zshrc so I'm going to grab that paste it here hit enter and that restarts the shell all right so interestingly enough I have a parse error in my zsh file which I've not seen before so I'm going to fix that with you live so I opened up my zshrc file and it said the problem is on 133 so let's take a look here it is cond initialize okay so I don't know why that happened but I'm just going to put it right there and now all of a sudden it should work so I'll save it and let's continue so I saved let's try it again all right that time it worked great so now it should work so let's just type fabric DH and there we go so it worked that's how to install it now let me show you how to use it the first thing you you need to do is give it your gp4 API key and if you don't want to use gp4 there are other ways where you can use Claud you can use local models anything you want but that takes a little bit more effort so we're going to type fabric D- setup and then I'm going to hit enter so I created a new API key I'm going to paste it in I am going to revoke this key before publishing the video hit enter and if we have a CLA API key it asks for that if we don't we're just going to hit enter and then it's updating all the patterns so it's downloading the actual prompts that we're going to be using and everything got added successfully but we have to restart our terminal to use them so back to this and we're going to do Source users Matthew Burman zshrc all right now fabric Dash just to make sure it works and it does and let me teach you about the terminology so here are the components we have the mill and the mill is optional but that's a server where you can make patterns available so if you wanted to open up a server to do that that's how you do it a pattern is the granular AI use cases aka the prompts the stitches are patterns that are put together in different ways to create even more sophisticated outcomes then we have the loom which is a client side app that can call the specific pattern but we're going to do everything CLI native so we're going to type fabric d-list just to see a list of the different patterns and look how many there are all of these so we have create logo create keynote extract references and again all of these are very tried and true prompt so let me actually show you that real quick so here is a prompt for analyze paper so identity and purpose you are a research paper analysis service output sections very well formatted very specific to reading and analyzing papers and then it gives output instructions so basically give the output and then we have the input right there okay so we have a bunch of those all right here so I have my video here that I posted yesterday this is the simple introduction to large language models I spent a lot of time on it so if you haven't seen it check it out if you want a basic introduction to llms and so we're going to grab the transcript right here we'll turn off the timestamps and I'm simply just going to highlight all of it copy switch back to terminal and we're going to do PB paste so PB paste and then you have the pipe fabric D- pattern extract wisdom and what this command is going to do is automatically take what's in my clipboard and then extract all the wisdom from it so let's see what it does now it's going to be a little slow cuz it's using GPT 4 but imagine if we were using grock with this that would be super fast and they're making incredible progress on this they're coming out with new features new prompts also known as patterns I believe they're coming out with a user interface to make the entire thing more easy to install for everybody now if we wanted to see the output while it was going we would just use dh- stream as a command and we'll actually see the output as it's going but there we go okay so it took my video this video video provides a comprehensive introduction to AI great ideas llm so it talks about all these different ideas that I put forth in that video it takes quotes from it these are direct quotes habits that are found in the video facts from the video references and recommendations on how to follow up based on all of this knowledge so very very cool extract wisdom so now I wanted to actually analyze my claims and stream it so let's go ahead and hit enter all right here we go claim one Ai and llms have revolutionized various Industries and here's the evidence and claim two so it actually extracts all the claims from my video and either supports or refutes them or both so and then it gives me an overall score so the lowest claim score is a c the highest claim score is a b interesting so I think this is awesome check it out I'll drop all the links in the description below if you like this video please consider giving a like And subscribe and I'll see you in the next one

Channel: Matthew Berman

Views: 133,332

Keywords:

Id: lEXd6TXPw7E

Length: 7min 3sec (423 seconds)

Published: Mon Mar 11 2024