PromptFlow: Create LLM Apps In SECONDS with NO Code FOR FREE! (Installation Guide)

[Music] remember flow wise it was an application that can easily build large language model apps within a couple of minutes it's an open source project that has revolutionized low code llm apps while there is another likewise project called prompt flow backed by the one and only Microsoft it's an open source project that is able to build high quality large language model apps from prototyping testing to production as well as through development and monitoring you're able to harness this application by creating your own workflow or by utilizing ready-made flow templates that you can utilize off their template documentation now the amazing part about prom flow is that you're able to create flows that link large language models prompts python code and other tools together in executable workflow which can be edited afterwards by others and implemented into your own workflow and this gives it a more Dynamic attribute compared to flowwise now you're able to debug and iterate gear flows also by evaluating them and calculating the quality and performance metrics with larger data sets you're also able to integrate this into your own Ci or CD system which can be Deployable now these are some of the handful of things that you can actually do with prom flow and it's something that we're going to cover throughout today's video so with that thought let's get straight into it and cover what prompt flow is actually able to do hey what is up guys welcome back to another YouTube video at the world of AI as we talked about the start we're going to be taking a look at prom flow more in depth talking a little bit more about how you can install it as well as taking a look as to what you can actually do with prom flow now we talked about this many times on this channel about flow wise in which you're able to create LM apps in seconds and this is by not having any code involved but this new application called prom flow is in a way way more dynamic as it gives you more features and attributes to creating large language model applications fairly easily with its documentation as well as its application you're able to create much more Concepts as well as applications with more Dynamic features and this is something that we're going to take a look throughout today's video as you go more in depth with prompt flow if you would like to access our private Discord in which you can get exclusive features Partnerships subscriptions to AI applications and so much more definitely take a look look at this link in the description below if you guys would definitely want to follow World debate I you'll be able to get the most latest AI content over here in terms of news so definitely give this a follow if you guys haven't and lastly make sure you guys subscribe turnification Bell like this video and check out our previous videos so you can stay up to date with the latest AI news so with that thought let's get right back into it so what is prom flow well prompt flow is basically a comprehensive Suite of development tools which they have stated and it's designed to basically simplify and optimize the complete life cycle of creating low code AI large language model applications and this is basically an Innovative platform just like how flowwise is but it's actually backed by a big Corporation such as Microsoft which redefines the way as to how you take approaches with large language model application development and this is by offering different types of features that float wise is actually not able to offer in this case you're able to create flows that link different types of large language models not just a list of large language models that they offer but a whole entire category of lists that flowwise actually doesn't you're able to link different size of plot prompts different size of python code and other tools together to make it an executable application now this is something that we're going to cover throughout the video as to how you can do it and how you can get started with this now some of the key features I want to talk about is that it's an intimate development cycle streamlining process you're able to take prompt flow which serves as this full-on encompassing solution and it caters to like every phase of a large language model application development this is something that they have a lot of documentation as to how it works but in summary it empowers you to like navigate to the entire process by focusing on each like step of creation creating your AI application and this is by making ensure that there's different types of prototyping to rigorous testing as well as different performance metrics that are used to finally produce and deploy your AI application now guys another cool feature is its executable workflow creation and this is something that we mentioned before in which you're able to seamlessly link large language models prompts python code and various other tools now this is something that will make this tool so much better in other ways as it gives you more capabilities in creating these different apps with different tools you have efficient debugging and iteration where it effortlessly uses prompt flow to deal with interactions while involving different large language models to solve your errors and code the performance evaluation feature with large data set is something that is very crucial with prom flow it is optimized to handle substantial data sets which allows you as a user to assess the quality and the performance of your workflow quite accurately you're able to also integrate this with CI and CD and it basically integrates the testing and evaluation into continuous integration as well as continuous development systems lastly you're able to also access this off the cloud version and this is something that we'll take a look at lastly with Azure AI you're able to take a look at prom flow on the cloud with this application and it's something which is highly recommended recommended because if you do not have the right understanding to install this code you can definitely just take a look at it on the cloud version and easily integrate different types of apps with this new Cloud version so with that thought let's get to the next step of the video and getting started with prompt Flow by installing it locally so now let's get to the next step of the video will now showcase how to get started with prompt Flow by installing it locally so there's a couple things that you will need beforehand you need to make sure that you have Visual Studio code installed as we're going to be working with prom flow with Visual Studio code you need to make sure that you have conda as well as python so once you have these three things installed we can get started so they have demonstrated what you need to do step by step you need to first start off with your Dev environment and this is by creating the python environment with conda so you need to make sure that you have python the latest version that's above 3.10 once you have that done you need to make sure that you start off by creating the python environment and then activating it and once you have done that you need to start off by installing the dependencies that are required for this so This Is by doing pip install prompt flow and once you have put this code in you need to check if that if you have installed the right version as well as it should promptly show the version 0.103 and this is by putting in this command and this is by confirming that you have this version installed of the installation now what you need to do next is understand what's a flow and it gives you a demonstration as to what you need to do in doing this you can get started by getting the flow sample and this is by cloning the repository of the GitHub repository that you can access with the link in the description below this is about copying this link or accessing it from this documentation by doing this command git clone now I forgot to mention that you need to have git and this is going to be used to clone the repository onto your desktop now once this is done you need to understand the flow directory in which you are going to be working with the different apps that you're creating and in this it shows that a flow directory is a directory that contains the all the contents of the flow the structure of the flow folder as well in this case it gives a good representation as to what the files are and how you can manage and work with this so in this case it says in order to run the specific flow you'll need to install the requirements first and this is by installing the requirements to run the flow and once this is done it gives you a demonstration as to understand the flow yaml file and this is something that this strikes or describes the drag and this is where you can basically work around with the flow in Visual Studio code so it also gives you an extension as to visualize the flow and this is fairly easily at the top of Visual Studio code you can search up for install prom flow CLI and this will give you the extension in which you can play around with the different types of flows and it gives you a representation is to choosing the python file that is needed for this by making sure that you check that it has passed the installation process and from this you're able to open the tag which is the drag visualization in vs code and this is by opening the flow.dag.eml file and this is open through Visual Studio code and you can see over here it's a good representation as to how the flow looks in which you can create different inputs different links of large language models as well as different applications now I'm not going to be showing this because there's just too much nitty-gritty things about this so I'm just gonna just give you a good representation as to how you can do it and give you a good demo as to Showcase what has been created with the flow or prompt flow sorry and in this case it gives you a good representation on deploying and testing your flow into different uh workflows so in this case they have a lot of tutorials on each step so I highly recommend that you check this out so you can get a good understanding as to what is being created and how it is actually being created with different videos so with this thought I will leave all the links in the description below and that's basically it for installing your own workflow to have a good representation on testing it and there's different types of next steps in which you can work towards and learn more with their documentation now guys one thing I really want to mention is how they have done a really great job in creating different tutorials for creating different templates and workflows you can see that you can get started with prompt flow in which they have a step-by-step documentation on how to install it this is obviously something that we're going to cover so don't worry about that but some of the other examples in creating different flows you have connection management flows they have a template in which you can create a a chat with PDF application you're able to do this to test the valuation experimentation run prom flow and Azure AI working with different connections and tools as well as setting up a different template for flow run management and a zero AI they have a lot of other different types of samples in which you can access off of GitHub itself there's different templates that are already made and which you can deploy with different paths so in this case you can deploy it right away with the docker for example or through Azure AI you have different types of flows that are already created you have an autonomous AI agent where this flow is showcasing how to construct an auto GPT agent with prompt flow to autonomously figure out how to apply the given functions to solve the goal this is just one example you have another example of different types of apps that are used as well as these templates in which you can create different types of apps there's so many of them so I highly recommend that you check this repo out in which you can have these templates and how to access them with prom flow it's fairly easy they give a good demonstration on how you can Implement these flows into the application once you have it installed another cool thing is if you want to incorporate other tools you can easily do so by going on to the how-to guide and it gives you a demonstration on incorporating your custom tools now we know that it provides the three basic tools as well as the other tools in which you can have a vector database lookup so in this case if you want to incorporate other tools you can just go to the Custom Tool package incorporate the prerequisites install and create the Custom Tool packages and basically put them into a folder within this so that you can launch it and build it Alpha vs code this is something that you can definitely do quite easily with this documentation so if you're interested I highly recommend you check out the how-to guides on doing this one demo I want to highlight is how they were able to create this chat with PDF application in this case they're able to upload a fault and create a chat bot for it so they have done this by installing the prerequisites by making sure that they configurate problem flow for vs code and this is something that we talked about previously in showcasing how you're able to integrate this through Visual Studio code by installing the extension once you have done that you're able to consult the chatbot chat underscore with PDF by utilizing the template that they already have in this case they upload that template and it gives you a good demonstration as to breaking down what is happening and how the documentations are built into the chatbot in this case they give you how to take a look at the chatbot and run it by your console app and from this it gives you what you need to upload in terms of the API keys in this case they have an open AI API game in which you can build your chatbot off of you're also able to integrate this with different large language models so don't think that is just restricted to open ai's GPT models in this case it also gives you implementation of each step so it tells you down the downloaded PDF file will be stored into a temp folder so in this case you can upload whatever temp file and at the end you're able to see that it gives you how the flow looks like in visual vs code and then you can deploy it and it will obviously look as look like the flow that we saw at the start where you're able to chat with different types of PDFs it's a really easy way to deploy this it's just a lot of coding so this might be hard for the basic uh newbie in which who doesn't have the right coding for like knowledge in doing this but if you're able to follow along with the description as well as the tutorials it's very easy to do so it's just implementing certain types of code within certain lines throughout each step process and if you're able to do and manage that you're able to create such powerful AI large language model applications and this is something that is better than flowwise in my opinion as you're able to have more flexibility and more capabilities in creating these applications so I highly recommend that you check this out and you got some sort of value throughout today's video now this is a game changer as this toolkit of prompt flow is able to build high quality large language model based applications so I'll leave all the links in the description below but thank you guys so much for watching make sure you check out the patreon page if you want to access our private Discord if you would like to follow world of AI on Twitter so you can access the latest news I'll leave this link in the description below so you can do that if you guys want to check out flowwise this video I made on how to install it I highly recommend that you do so if you do not want to install prom flow and you don't want to basically do that coding as this is another good option in which you can create large language model apps through a smart user-friendly UI lastly make sure you guys subscribed turn notification Bell like this video and check out our previous videos so you can stay up to date with the latest AI news but with that thought guys that's basically it for today's video thank you guys so much for watching have an amazing day and I'll see you guys fairly shortly peace out fellas

Channel: WorldofAI

Views: 12,169

Keywords: langchain tutorial, flowise ai, langchain agents, PromptFlow, Microsoft, ai, microsoft ai, Create llm apps, flowise, open-source project, langchain.js, llm applications, gen apps, llm apps, prompt chaining, llm chains, llm application development, prompt engineering, agents, chaining, create llm aps, create llm apps with no code, create ai apps without coding, llm app development, langchain, no-code langchain apps, opensource, opensource ai tools, ai app development tutorial

Id: nveTxm9TS1Q

Length: 15min 35sec (935 seconds)

Published: Fri Sep 08 2023