FREE LocalAI Open Source API OpenAI Alternative!🤖 Chat, Image, Voice Generation GUI TTS Embeddings!

What's going on guys Josh poock here and in today's video we are going to take a look at setting up local AI the free open source alternative to open aai Claude and others we're going to show you how you can self-host this for free this tool allows you to generate text audio video images Voice cloning and a bunch of other cool stuff you'll be able to access all this in a web goey or have the option to run it in an

API so you can integrate it with your apps or anything that you want let's dive right into it before I show you guys how to set it up let's just quickly go over the features so text generation with gpts text to audio audio to text image generation with stable diffusion open AI like tools API embedding generation for Vector databases constrained grammars download models directly from hugging face Vision API reranking API P2P interace facing and then integrated web uui and this Project's

Been around for a while it has over 25k stars on GitHub and they actually now have a really simple setup so all you're going to have to do is have a Docker installed on your system if you don't have Docker go to doer. get that set up and running and then you can either run the installer script like so now I was having a couple issues with this um I don't know exactly why I didn't troubleshoot too hard but for me

It was just easier to run this right here okay and you'll see here that there is different options for images so you can run this one right here for CPU uh if you have an Nvidia GPU could run this one without preconfigured models you could just run this and then without pre-configured models for NVIDIA GPU you could run this okay cuz when you run this it's going to install some preconfigured models and you're going to need about probably like I don't

Know the exact number but it is a pretty big um amount of space you'll need maybe 10 to 20 gab just to be safe I know there's about four or five three to five different models I believe that's going to pull up all the docker containers and install all the different models here once you install the models and get this whole thing set up you can go to Local Host port 8080 to access local AI all right so here we are

At our local AI instance here you'll be able to see the documentation as well so I would recommend you know if you're going to dive deep into local AI check out the documentation they have some pretty good uh stuff here covering you know getting set up all the different features cuz like I said there are a lot of different features we won't be able to go into each one of them so in depth in today's video and there's different Integrations Advanced

Features Etc so definitely check that out now on the homepage is where you'll see your different models so here we got eight installed models these are the default ones that come so we got gp4 uh GPT 40 Gina ranker stable diffusion text embedding Ada TTS one whisper one back laava right here so um you can delete specific ones if needed and over in models right here you'll be able to install different ones so as you can see available 536 models from

One repository up here you can filter by TTS models you can filter by image generation models text generation multimodal embedding models rankers audio trans descriptions all right so you can see a bunch of different models here we got quen coder you know so many models you can really choose from some of them have uh logos here some of them don't you'll be able to see the repository the license you can go info here and it will pull up some information on

The model you can see the links so hugging face right here different tags and if you want to install the model it's very simple you just click install and it will get going so you have the option to install so many different models like I said um and then you can so chat with the models over here so in chat you'll have a nice chat gooey right here so you can go up here and select different models you want to chat

With you could do different ones of course that appear here you can set a system prompt if need be you can set an API key right here and you could clear chat I can say tell me a joke all right and it said why don't scientists trust Adam because they make up everything ha so funny okay so there is the chat you can of course change different model models here download more models as well of course so if you come over

To the dogs here you can of course chat with you know local AI in the guy the web guey like I just showed you but you can also use the API endpoints to send API requests to local AI in your applications or just in general HTTP requests to uh chat with the model do chat completions Etc I'll show you that in just a second but while I'm here on the Integrations page you can see we got anything llm we got code

GPT flow wise Lang chain uh llm stack a bunch of different Integrations so that's pretty cool so if we go to Features here and text generation you can see the API reference so chat completions right here you could also reference the open AI docs right here but you could see we could make a curl request to Local Host 880 uh check completions right here and you could select your model you could change your message right here so roll you uh content

Right here and your prompt temperature right here you have additional parameters like top he top K Max tokens you could do edit completions completions list models a bunch of different stuff so check out the docs all right next is generate images so if we go to generate images here you could select um your text to image model so we would do stable diffusion and then you can do something like monkey playing Ino and then send okay so here's our image uh

Obviously not perfect the monkeys are kind of deformed and I don't know what type of bingo game this is but of course you could try different models do different prompts of course you know improve that this is just a test I did get a better one before when I was doing tests and of course if you reference the docks here you can see the image generation as well as um the different API end points right here that you could use to

Generate images okay you can also come over to API right here of course and you'll see the Swagger API right here so you can see all the different API endpoints for pretty much everything within the app all right next is text to speech SL audio so we can go ahead and prompt it right now and it will generate that into speech format first off I'm going to select the model and put that to TTS all right and we'll just say yo

Subscribe to Josh okay and then we can click on download results save that file and we can open it right here and take a listen you subscribe to Josh okay so pretty cool of course with any of these you can use different models then let's go over here to talk all right so here you can change different models like llm model we'll just do gp4 whisper model we can do whisper TTS we'll do TTS and then we can go ahead and

Click talk now I don't know if it will be able to pick this up just because I'm recording right now let's go ahead and try hey how do you make a cupcake in explain it to me in 30 seconds okay so now it's processing the audio that we just spoke all right so now you can see that it picked it up um well let's go ahead and try okay how do you make a cupcake and explain it to me in 30

Seconds all right so I picked up our audio now it is generating the response all right and as you can see we have our output right here so I'm going to go ahead and play some of it in 30 seconds here's a simple guide to make cupcakes one preheat oven to 350° f 175° c then mix 1 D 1/4 cups allpurpose flour 1/ 2 teaspoon baking powder 1/ 2 teaspoon baking soda and a pinch of salt in a bowl pre all

Right I think you guys get the point so that is the talk feature all right and then there is swarm up here so distributed inference with P2P so distribute computation by sharing and load balancing instances or share uh sharding model weights so you can read more about this feature right here in the docks and how you can actually use it right here you would actually have to start local AI uh right here using these flags so P2P and Federated so I'm

Not using it right now so I'm not going to cover that but this is more a little bit more advanced stuff if you want to check that out but that's pretty much it guys for local AI just to summarize guys this is a powerful UI interface that allows you to download different models from different types such as TTS image gen text gen embeddings reranking you can do TTS of course talk generate images all that good stuff and you can access all

These features within a nice API similar to what you would use with open AI so you can connect these into your apps do a bunch of different stuff with this or just use the web gooey let me know what your thoughts are guys about local AI in the comments Down Below have you used it before you know you use it for some really cool features that we didn't cover in this video do you like it or do you know a better

Alternative let me know what your thoughts are down below guys other than that if you're new to the channel and you got some value here then like the video comment down below and subscribe to stay up to date with the uploads We Do videos all the time on AI Marketing sales business growth Etc also too guys if you haven't already joined our free community stri community.com I'll leave a link down below for our free Facebook group and Discord Channel then also

Too guys if you run a business and you need help with marketing Sales Systems implementing ai ai agents appointment Setters Ai call centers into your business to streamline your sales process and Lead flow book a call down below at executiv strip.com /a and we can see if it's a fit other than that guys I will see you in the next video keep hustling keep grinding and of course guys accelerate your stride take care are you tired of pouring thousands of dollars

Into appointment Setters only to watch leads slip away imagine having a team of elite sales agents booking qualified appointments for you around the clock no more wasted time on training no more frustration with performance and no more draining your budget on inconsistent and expensive call centers introducing stride agents AI powered appointment Setters that work 24/7 never get tired and book appointments while you sleep trained on thousands of successful conversations our AI agents outperform human teams at just on10th of the cost

Join the ranks of businesses that doubled their appointments and booking rates in just a matter of weeks don't get left behind in the AI Revolution visit stride agents.com now and transform your entire sales process with cuttingedge AI technology it's time to accelerate your stride with AI agents