RestAI: AI as a Service! Create Software, Rag, SQL, and Vision Apps!

document knowledge base into this rag Gap so with this UI you're going to be able to build all of this and you're going to be able to have this functional app which you can ask questions with so what I've done is that I gave it the system prompt which is stating that you are a digital assistant answer the question about the following context never invert an answer if you don't know the answer just say you don't know if you don't understand the question just say you don't understand so what I asked was that the hard hard thing about hard things explain this PDF I basically just said that this is a book and I want you to explain it and then it was able to give me this answer where it stated that the hard things about hard things is a book that discusses challenges faced in management and self up situations it goes on to give me the best information that is needed to retrieve and basically fulfill the query that I provided it and you can see that based off of its rag system it was able to retrieve it from the par section of the document that was uploaded and you can see that it does a great job in detailing all the different sections of the book and you're going to be able to get the relevant answer easily sorry for being repetitive but this month we had insane Partnerships with big companies giving out subscriptions to AI tools completely for free these are tools that will streamline your business's growth and improve your efficiency just being a patreon this past month you were given access to six six paid subscriptions completely for free not only do you access these subscriptions but you gain the ability for Consulting networking collaborating with the community as well as with myself you get access to daily AI news resources giveaways and so much more if you're interested check out the patreon link in the description below to gain access to these benefits now throughout today's video we're going to be taking a look at rest AI further in detail showcasing what you can do with AI as a SAS as this is is definitely a great tool that will be helping you create prototypes applications rag apps and such forth so with that thought guys stay tuned and let's get straight into the video if you would like to book a one-on-one with me where you can access my Consulting Services where I can help you grow your business or basically give you a lot of different types of solutions with AI definitely take a look at the calendar Link in the description below hey what is up guys welcome back back to another YouTube video at the world of batei in today's video we're going to be taking a look at rest AI this is a comprehensive platform that's offering a unified user interface for various air related tasks it's going to be helping you create different projects as well as different apps prototypes as well as utilizing Advanced features like text tosql and multimodal inference as a service now this platform is aiming to simplify the Integrations of AI capabilities into project S without requiring extensive technical expertise it's making your life easier to create these different SAS products or AI products and it's by providing accessible tools and services through rest AI so you can see that there's various features you have projects there are multiple types of Agents projects and each of them have their own features you have a rag project you have a rag SQL inference as well as Vision you can actually take a look at all of these different demos so if you are to go over to the rest AI website you're going to have to sign in the password for it is demo and the username is also demo and you're going to be able to access all these different projects that are available right now and you're going to be able to test out the different prototypes that have been created using rest AI now going back as another feature which is users you are going to be able to represent a user of the system it's used for authentication and authorization basic B like you would say basic authorization for accessing certain portals you have large language models where you can support any public large language models that are supported by llama index or any local large language model supported by oama you have the utilization of vram as well as apis and the front end that is available through rest AI front end and this is something that we're obviously going to be taking a look at as we go further into the video showcase how you can use it and how you can develop certain things so now let's take a look at some of the project types because this will give you a better idea as to what you can do with rest AI firstly we're going to take a look at rag it offers obviously diverse types of things such as embedding models like chroma and redus and this is for semantic representation you have retrieval functionality for assessing embeddings as well as the quality of it you can also even refine the search results with options like cobbert as well as other large language models and this is quite versatile cuz you can work with data loaders and it will have you operate in a sandbox mode which is going to provide default answers for queries lacking embeddings so this would be really useful for chat Bots where you can inter integrate the rag into that chatbot and it's going to be helping you in this simplified matter and it will make sure that you're going to be able to utilize it through this sandbox next up you have rag SQL this is where it's going to streamline the database interactions and this will enable you as a user to supply your own SQL or posters SQL connection and it's going to be allowing the system to automatically crawl the database and it's going to utilize the tables the custom columns you have column names or even translate your user queries into SQL commands this is all doable within the project type that rest AI provides it can also even generate responses based off the retrieval data which is really cool and it will streamline and simplify the process of querying databases and extracting relevant information next up you have inference which is referring to the process of drawing conclusions or making predictions based on the avilable evidence as well as the data that is uploaded through rest AI it's going to be able to do it quite fast and this is one of the great features of this project next up you have vision and this also utilizes lava which is really cool to see now with rest the ey it's going to be able to facilitate text to image you have image to text as well as image to image now in this case the generation is using local stable diffusion models as well as Dolly e and this is based off of different P romps that are being boosted through an internal large language model and this is for enhanced details obviously so that you're going to be able to get characteristics like changing the color of an image as well as adding different generation to the image now for image to test tasks it's going to be supporting lava as well as back lava and that's by default so you're can to be able to integrate with the from Vision models that are already integrated through the project list of rest AI now for image to image task rest AI actually has provided support for instant ID as well as quen's Vision large language model and this these are some of the models that they have provided as default now what they have stated is that you're going to be able to deploy and use any large neish model that is supported by Ama or llama index so see if you are to install this locally and you have these models that are supported by these two Frameworks you're going to be able to integrate it with rest AI so you get various other specifications or Integrations of large nage models within rest AI now guys there is an installation method it's super easy you clone the repository once you go into the directory of the repository that you cloned you can then install it using Petry once you have done that they showcase how you can deploy it and it will be then on your Local Host which you can access fairly easily now I'm not going to be showcasing that CU they already have a cloud version that you can try out this is obviously a demo so you're not going to be able to create projects with it but it's going to be able to get you a good grasp as to what you can do with it now as I mentioned there is five different project types you have inference rag vision and then you also have rag SQL and you can see that there's various other large language models that are supported with this like Gemma as well as mistro so for example if we are to take a look at the demo 5 which is focusing on inference you can check the details will showcase the large langage model and it will show you the system prompt where it says always end your answers with beep beep and then you can go over to the question you can obviously ask it questions so you can resume the text in this case this is an example already here which showcases the Su system summary where is summary the summary the following so I think it meant to say summarize the following text in 30 words or less don't do not invent anything stick to the provided text if there isn't enough information answer saying that now we can see that the question is termo has engulfed the galactic Republic the taxation trade rumor routes to outline star systems is in dispute I'm going to just simply ask this question and I'm going to see what sort of answer I get out of it and looks like based off the provided context that I gave it it was able to give me a good answer which solves the like query that I basically asked it and this is something that will showcase how it's going to be able to refer to the process of drawing conclusions or making predictions based off the evidence that I gave it and it was able to do a great job based off of what I've provided now if we are to go back to the project list we're going to be able to see that there's other demos such as rag where you can utilize the raged or the mistal model to showcas how it's going to be able to utilize rag you can take a look at Vision which will'll use lava 16 so in this case you can take a look at the details which will showcase the model you can click on Vision the prompt is something that you can obviously upload as well or you can utilize certain things such as the templates that they are already provided so you can generate stable diffusion local which is using stable diffusion to generate a picture of a happy cat so if I ask this it's going to be able to answer within a couple seconds now I wouldn't say this is a happy cat but it is a cat so I'll give you that but you can get a gist of it it's going to be able to answer your queries based off of vision inputs and there's various other options such as rag SQL I'm not going to be going into that cuz I don't have any queries or any databases uploaded I believe that they already have certain things uploaded so you'll be able to check that out but if you are interested take a look at some of these project types you're also able to create your own project so this will be really cool to see as to what you build with it based off of the different templates that are provided they're obviously going to be elevating this as time goes further so definitely stay tuned but that's basically a gist of today's project our product which is AI as a service it'll showcase how you can build AI products and how you're going to be able to provide this service to other domains this is something that is fully open source so you can get started with it right away and you can utilize it in various use cases so with that thought guys I hope you found this video to be very valuable make sure you check out the patreon page if you haven't already this is a great way for you to stay up to date with the latest AI news as well as different AI subscriptions that are completely for free make sure you follow us on Twitter this is a great way for you to stay up to date with the latest AI news I'm always posting different AI content over here make sure you follow us and subscribe on YouTube make sure you check out our previous videos so you can stay up to date with the latest AI news and I would highly recommend that you turn on notification Bell because I'm always posting different sorts of content that you will definitely benefit from so with that thought guys have an amazing day spread positivity and I'll see you guys fairly shortly peace out was

Channel: WorldofAI

Views: 7,315

Keywords: artificial intelligence, ai as a service, restai, create software, rag apps, sql, taipy, create software with ai, create production ready apps, create ai agents, create production ready apps with ai, taipy python, chatdev, chat gpt, software development, communicative agents, copilot gpt, devopsgpt, coding with gpt-4, devops full course, autonomous ai agents, auto ai, autonomous ai, how to create softwares with ai, how to create softwares with chatgpt, llm, software automation

Id: aVYe4uBo1PI

Length: 13min 0sec (780 seconds)

Published: Sun Mar 31 2024