FREE big AGI Generative AI Powerhouse Suite!🤖 Open Source Any LLM AI Agent - Chat, Browsing & More!

What's going on guys Josh poock here and on this channel we've covered many different AI tools AI open source tools and so many different kinds but what if we could have one open- source tool that combines a bunch of different AI Solutions into one well that's the tool that I'm going to be showcasing in today's video it is called Big AGI with AGI on the cusp of being invented and being in the Forefront we have this tool right here which which

Is a generative AI Suite powered by state-ofthe-art models providing Advanced AI AGI functions it features AI personas AGI functions multimodal chats text to image voice response streaming code highlighting and execution PDF import presets for developers and much more you can deploy on premore in the cloud they claim to be the most Innovative and productive generative AI Suite I'm going to show you how you can self-host it in the next few minutes let's Dive Right In into it all right guys so

All links as always will be in the description down below the link to their website as well as the gith Hub repo so you can self Host this for yourself um here's the website you know here's kind of a glimpse of what it's going to look like and they have a Discord Channel just to quickly go over some stuff tailored AI with all these different models right here as you can see some mistol open AI open router local AI models which

We just covered local AI on this channel power for ux private so chats are stored in your own browser uh Advanced fun call AGI mobile UI voice UI and then with personas so this is really cool I'll get into that in just a second and then more as well all right so they have some pretty cool tools right here they have chat they have call uh which we'll dive into in just a second and they also have beam so you may

Be wondering what is beam well as users become more familiar with AI uh language models the limitations of the models can become increasingly apparent and frustrating leading to a desire for more intelligent and controllable AI interactions beam a groundbreaking chat modality in big AGI addresses these needs by enabling users to easily enable multiple models simultaneously fuse their best responses and Achieve better output wiser decisions with lower risk and reduced AI hallucination so they claim this leads to better brainstorming and idea

Generation abilities more informed decisions with more breadth of exploration to dilute and reduce the incidents of llm hallucinations and then higher stake outputs that far exceed cost trade-offs the when engaging with beam chat the user sends a message to beam using Control Plus enter instead of just a single llm this is presented in the diagram below so we can see you could send off a beam right here it could go to open AI anthropic uh Google Gemini and mistol all at

Once and that's what beam does and then step two will be the AI merge so it combines responses into a coherent best answer under the user guidance right here so fuse response to user so pretty interesting of course read more about that feature in right here I'll leave this link in the description down below as well they have a full blog post on it so there's the documentation right here I'll leave that link down below and there's some Advanced features and

You know going over different models and such we'll come back to some of this in just a second let's go ahead and talk about getting this set up so the GitHub repo will be linked down below of course and you're going to go here and you're simply just going to go to installation right here I'll actually leave the direct link down below so this is the installation guide it's very simple if you've installed any sort of GitHub repo before you're simply

Just going to get clone the repo and then CD into big AGI I think you guys have seen me do this enough times I'm just going to show you right here but obviously you're going to open up your terminal and do so and then you're going to run npm install and then npm run Dev okay that's going to get the uh instance running on Local Host Port 3000 oh I'm just noticing that it is using trpc here so that's cool anyways

So once you do that of course you're going to go to Local Host 3000 look like this okay so you will see that we have the chat right here now I'm just going to move this over just so you can see over here on this side we have chat um um and we also have chat so Persona replies write so append a message draw so AI image generation and then reason plus Rea uh reason plus act so answer questions in multiple

Steps I'll show you that too in just a second we also have beam right here so this is what we talked about and then we have call so quick call regarding this chat so pretty cool stuff here we got the AI Persona so we have like default developer scientist Catalyst executive custom person Persona Creator so right here it says start here so if you click here it's going to show different examples like help me plan a trip to Japan what is

The meaning of life typical examples you would see prompt right here is going to show you the system prompt so you could customize this you're an assistant right here it says knowledge cut off this is the default one current date Um Mermaid rendering enabled plant uml rendering enabled SVG and markdown rendering enabled data press presentation prefer tables Auto column so you can of course customize this if need be you can also change the titles right here and what are be what

AI personas are going to be displayed right here over up here we have the um llms so you can choose an llm right here I'll show you how to configure these in just a second and then here is the AI personas as well so you could choose default custom whatever the case may be over here we have the chats so you could search for stuff in your chat you could create a new chat you could delete a chat import chats export

Chats delete chat and then use folders so you could create new folders like test like so save that you can create another folder like folder okay think you get the point you could add a chat to a folder like I add this to folder and now we see the little pink folder icon here we go up here to the models and then click on models this will enable us to to configure our models so we could go anthropic Gemini open AI

We'll just do open AI for Simplicity sake and I'm just going to paste in my API key right here gp40 so you can enable different ones so 01 mini 01 preview real time voice um gp24 you know all the different ones you really want you can see you can refresh the models you go all services right so you got a bunch of different options here you can also Advanced here which allows you to put an API endpoint so you could use

Helicone cloudflare organization ID a helicone key here observability moderation on or off okay once you have that you can go ahead and actually use your model so right now we're using gpg 40 I can just go ahead say hi you'll see that we have that little gift there chatting and hello how can I assist you now you'll see here we have different options right here if we do different uh forward slashes so we can do forward slash Beam for slash draw

All right you could drop files um so let's go ahead and do forward SL beam SL beam high and then we can see we get welcome to Beam Beam mode so beam is uh chat modality blah blah blah explains it so we can go ahead do next okay it's explaining it here go next and then we can go ahead and do start so this is actually where you can select your different models so we could do GPT 40 we could do

Claude Sonet okay so of course you're going to want multiple llms already saved in here so I'm going to post in my API key for anthropic all right so now I can do gb40 claw son 3.5 now you can easily remove models like so or you can add more models to the beam you could do like eight times four times two times by default you could add you know a bunch like let's go to eight here yeah so eight is the

Max I'm just going to do two ryp 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 centers that work 24/7 never get tired and book appointments while you sleep trained on thousands of successful

Conversations our AI agents El performed human teams at just oneth 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 and you can see down here we have our merge model right here so there's fuse guided compare and custom and then

We got merge model selection right here I'm going to go ahead and test this gener we can see generated right here do generate for cloud Sonet okay so it looks like Cloud Sonet was a little bit um longer I noticed you wrote high beam which seems like it might be a typo would you like me to blah blah blah okay so now we can see that we have access to the merge model now so you could select a different merge model

Whatever one you want you could fuse them so a add Fusion AI combines conversation details and ideas into 2 one clear comprehensive answer there's guided so add checklist choose between options extracted by AI from the replies and the model will combine your selections into a single answer there's compare so analyzes and Compares AI responses offering a structured framework to support your response choice or custom toine your own diffusion prompt so pretty um a bunch of different you know varieties that you

Can choose from here we'll just do a basic fusion and of course I just said high so we're going to get some basic stuff um combine response hello it looks like he may have intended to say hi or possibly had a typo with high beam if you have any questions or need assistant feel free to let me know how I can help Okay so this one said how can I help both of them asked let me know how I can help

That's pretty standard um but I think you get the point okay now I'm going to do another beam just a little uh you know one that has a little bit more um you know content in it and also too you can see right here this is pretty cool it actually has um something you hover here and it shows you the estimated um so we can see available tokens message right here model max tokens this message contains about 16 tokens history so

Max response right there chat turn cost so you can see all the different statistics for you know how the cost is um and you know how much tokens you're using Etc but let's go ahead and just beam this question right here I'm going to use the same models just click Start here and you can see both of them are generating like imagine if we were using eight different models right here all right so as you can see we got you know

A paragraph from each different llm right here so we can see this one's they're both talking about the history so let's go ahead just do a basic F Fusion right here so I'm going to fuse them and we're fusing them using gpg 40 but imagine like you could add for example uh like Gro you could add mistro um you could add it would be cool if they had like perplexity too um you know just cuz that's really good for Powerhouse for

Research but anyways and then you could combine them with like any model you use 01 to combine them right so here's the combined response I'm not going to read it all but you could simply just copy you could use this message so now this is the message that we use from the beam now it's in our chat right so we're getting the Best of Both Worlds from gp40 as well as CLA on it you can use 01 with CLS on it

Right really get the best response and see what llm gives you the best response for a certain um question that's a it's a really good way to compare actually all right if we do forward slash you'll see that we also have draw react browse so we'll assist you uh we'll download the web page assist system clear and help all right let's try draw so drawing an apple give me a a dozen seconds while I draw an apple it's using dally right

Here and boom we got an Apple so we can uh click on show prompt and it will show the prompt so it actually creates a promp tier a vibr vibrantly colored ripe Apple glistening with the morning dew it skin is smooth and reflects the sunlight creating a beautiful spectrum of red yellow and greens so cool image here um it's actually nice and you can click open a new tab of course you could save this image it actually does a really good

Job um like look at those anyways I think you guys probably seen a million AI images but look at the water on that and everything anyways we can also click on on call so the call button I showed you right here we can click on this and okay so my browser I'm using uh zen zen browser uh which is like Firefox it doesn't support speech recognition issue for some reason so I'm just going to open it over here on my Brave

Browser a side note I just came to the new section here and this is Agi big AGI 1.16 it looks like they're building big AGI 2.0 you can apply for early access coming this fall um so they're actively developing it and I really I do like it like you know even just the the me like the gifts here like the the little rodent um it just kind of you know adds some flare to it I like the the style of this

App and uh you know they have some cool features but like I was saying if we go to voice right here this is where you can actually call Big AGI so you can call different personas um so we have the default developer scientist catalyst so if we click on scientist here we're getting a call from a scientist accept we can go allow mic before I actually do this I just noticed if you click up here you can see preferences so okay

Perfect there's a dark mode I like this app even more um then change voice okay so I already connected my um 11 Labs so you're going to have have to add your 11 Labs API key and see I did so so I got access to this is like my actual voice clone right here um you can change the gray UI voice call feedbacks if you click here and then click on this Cog icon you'll be able to access um the settings

Here preferences so you have chat AI models all these different options here appearance Zen minimal clean text size page width full medium um chat AI Auto diagrams that's kind of cool auto chat title Auto questioning Labs so performance so you can change these things on uh screenshot capture webcam capture cost messages um you know all these different options here so really cool that they have all these features without love is like a tree without blossoms so that's my voice with 11

Labs so um this where you can invoice you can add your 11 Labs API key uh language mic timeout speak responses and then select your assistant voice here you can see shortcuts here draw so this is where you can change like right now we're using Dolly 3 could also use Pria if you have a Pria API key and then tools right here so this is browsing so you have browsing um which you could configure um using Puppeteer right here using a

Web socket um I'm not going to cover all that right now but it does mention that in the docs which I was uh referring to earlier if you go over here to documentation you go to browsing you can see information on browsing functionality right here as well as react so questions answering with reasoning and actions you can read more about those two quote unquote Advanced features in the docs and you have Google search API so you can do that as well

To get some searching functionality with tools now I can actually chat with these models right now uh just because I'm using my mic on my phone right now to record but as you see it's not picking it up but you go ahead test it talk to some of these models have your little AI pen pal um also too if you go over here to personas personas is really cool because one you could just add a bunch of text like paste text

From tweets social media post whatever in here and create a Persona using like whatever Mall you want and you can go Advan prompts you can see what the prompts are so like Creator system prompt you can read this I'm not going to read through it all but analyze the text so you can see how they're creating these Pro uh Persona prompts just from their prompts which is pretty cool this give you ideas just in general and you know prompting um I

Always say like you always want to be kind of studying what a good prompt is because you'll never what even if it's a prompt in something like this this could help you if you're coding or doing whatever and seeing how good prompts are built cuz uh that's really one of the most important languages you can learn right now is prompt engineering anyway so here you can also do YouTube which is really nice so let's go go ahead to one of my

Favorite podcasts and we'll just go ahead Trump's mandate and extinction level event uh for federal bureaucracy we can just copy this link right here and go back here paste this in okay and we can go ahead it's going to use these different prompts right here we're going to use gpg 40 and click on create it's analyzing the transcript so generating you can see the bite's going up this may take one to two minutes so we can see it's working input text

There there's this kind of thing that happens in biology called Evolution and a lot of people think it's continuous process blah blah blah blah it's analyzing the transcript here so the speaker in the transcripts presents as an analytic and strategical uh thinker involved in professional field that bios system systematic evaluation and strategic disruption okay I didn't finish it all but oh wait yeah so it it went through the whole thing it went through all those different prompts uh to create the

Persona prompt which is this so you may now copy the text and use it as a custom prompt so you are strategic disruptor name not specified age mid 40s to 50s which is like you know you know this typic Persona of um Allin podcast occupation polit political analyst and strategic consultant cultural background Western likely American personality traits analytical strategic objective communication style metaphorical direct yet nuanced Eng engaging motivations uh efficiency and wow so imagine putting in any video you want if

Someone may be speaking in the way that you like to speak and getting a Persona that's this is actually very accurate to how this you know video was um just how they talk and uh specifically aspirations systemic reform uh fears and concerns stagnation potential negative outcomes core values resilience efficiency Innovation Etc so this is I think you guys get the point here you can go ahead and copy this prompt you can go ahead go back to your personas here or you

Could just use this like really anywhere obviously this is like you could use it in big AGI or you could use this in Claude or wherever the case is um I think you guys get the point here but there's a bunch of cool stuff that you can do in this tool like I said it's a pretty cool little AI Powerhouse here with some Cool Tools functions Etc you could definitely do a lot more than what I even showed in this video

I just wanted to give you a general sense let your imagination run wild with this but let me know what your thoughts are in the comments Down Below guys what cool things are you going to do with this tool have you used it already are you planning on using it what do you think of big AGI other than that guys if you're new to this channel we upload videos all the time on AI Marketing sales business growth Etc so if you

Like that type of content you got some about here smash the like button comment down below and subscribe to stay up to date with our uploads thank you so much for the recent 13k 20K on the way also too guys if you haven't already joined our free community Shri community.com I'll leave a link down below to our free Facebook group and Discord Channel and then also too guys if you haven't already joined our free community which I literally just made stride

AI Community you are definitely going to want to do this I'm putting my all into this community and I'm going to build it with you guys here I really want to uh interact with you guys get some give you guys more free value and just build a good cool Community here so I'm going to be making um a exclusive AI resource Vault behind the scenes stuff that I don't even post here on YouTube all the resources and more are going to

Be in this AI community so if you've ever got any sort of inkling of value from this channel there it's going to be tenfold in this community and you're going to get to interact with myself as well as other like-minded individuals all right so definitely check that out I'll leave a link down below and in the Ping comment and then also too guys if you run a business you need help with AI Marketing sales implementing AI agents cool callers call center

Into your business running paid ads whatever the case is optimizing your systems uh book call down below at executive.com slapp to speak to myself and my team and we can see if it's a fit or not anyways guys I hope you enjoyed this video have a great weekend guys keep hustling keep grinding I'll see you in the next one and accelerate your stride take care