This video was a lot of work to put together. Hope you enjoy and find it helpful. Please hit the like button to help promote the video. Also let me know if you'd like to see a tutorial on integrating Flowise with a specific platforms.
@arthuraquino83568 ай бұрын
An example using flowise and crewai would be very interesting to use in crew agents tools that are flowise flows.
@NileValley-TV8 ай бұрын
Thanks Leon, definitely appreciate it. getting ready to watch it right now. Question, when you make the chatbots and agents on Flowise and then put them on customers site, how do we guarantee continuity/persistence of the bot or agent on the customer's site. If you already did a video about that, please point me to it, thanks.?
@hosst8 ай бұрын
Your videos are actually useful, not just playing with AI and LLMs but things that be applied in real life and projects, thank you for this. I hope you create a Udemy course one day, I will buy it for sure
@TorstenHeissler8 ай бұрын
Done! Again a great video! Can't wait when Flowise goes "Self Learning GPTs" kzbin.info/www/bejne/hZ-0gpiNetytr9ksi=tSIIOFeWcrUV8QCL
8 ай бұрын
Easily the best content on flowise and really any LLM tool setup. Very grateful. 1.) Do you have any played courses or mentorship cohorts planned? 2.) Deployment and production are the hardest and least mentioned in the industry, Any info you make for deployments would be great. Loved your video for render.
@abedmalak_8 ай бұрын
editing this video must have been super tiring, we appreciate your hard work @Leon Van Zyl !!!!
@leonvanzyl8 ай бұрын
Hahaha, it took surprising long to script and edit! Thank you for noticing 😅
@abedmalak_8 ай бұрын
this is by far the best video you made!!!!! I learned the most from this!
@leonvanzyl8 ай бұрын
You're welcome! Trying to step up my game over here 💪
@wilschoy788 ай бұрын
I want to thank you a lot this is a lifesaver all flowise video and I was able to integrate the chatbot to a basic moodle instance by adding the html at the footer, now seeing this upsert by uploading a file inside moodle is my next challenge, keep this up bro without this I wont be able to made the moodle integration I will keep waiting for the next video. Thumbs up Leon!!!
@leonvanzyl8 ай бұрын
Thank you very much 😊
@MaliRasko8 ай бұрын
Amazing video Leon. Much appreciated. I was struggling with many of the concepts you presented here. I solved most of them eventually but it took me sooo much time. Your video still delivered many details that filled the gaps in my knowledge. What I am still struggling with is how to call an API from Flowise. I can see there is Request Post in "tools" section... but I was not able to execute it properly .. especially when I wanted to have a variable in the json body.
@leonvanzyl8 ай бұрын
Thank you VERY MUCH for the generous support. These super thanks make a massive difference and keeps this channel going 🙏
@abedmalak_8 ай бұрын
I was struggling with using upserts with APIs for 3 days and you solved it in 1 minute!! I love your youtube channel LEon!
@leonvanzyl8 ай бұрын
That is awesome to hear! Glad I could help
@julius_chun6 ай бұрын
I appreciate your work!! great preparation and teaching. this is incredible, thank you Leon. You are the single greatest source of knowledge of using flowise, which I am benefitting a lot from!
@leonvanzyl6 ай бұрын
Wow, thank you 😊
@j0hnc0nn0r-sec8 ай бұрын
Wish this guy did all the tech tutorials
@mohammd1007 ай бұрын
thanks Leon for this video I was waiting for it. Keep the Great work up
@rodrigorubio34984 ай бұрын
Mate. This is GOLD! thank you. Just what i was looking for, advanced options.
@leonvanzyl4 ай бұрын
You're welcome 🤗
@WayneBruton8 ай бұрын
Leon, really enjoyed this video, kinda saves me a lot of time in python. I am a strong believer of "Never reinvent the wheel". Still really looking forward to assistants with streaming.
@MariushAI8 ай бұрын
This video is amazing, I have been looking for a tool like Postman for API testing. Thanks a lot mate!
@leonvanzyl8 ай бұрын
You're welcome 🤗
@BadBite8 ай бұрын
This video is concise, complete and has given me lots of ideas, solved many doubts. Demanding work Leon and I appreciate it! Thank you. I would love a Streamlit tutorial as I work with python most of the time!
@leonvanzyl8 ай бұрын
Thank you!
@musumo19088 ай бұрын
The best video on flowise API config so far! You’re on fire! Would love to see something on custom js in FW…for lower coders…thx
@leonvanzyl8 ай бұрын
Thank you! Also, great suggestion.
@AssassinUK4 ай бұрын
Been waiting for THIS video for a long time. Thanks!
@msclt1982Ай бұрын
Absolutely loved this video! 🎉 The explanations were super clear, and the step-by-step guidance made everything so easy to follow. Thank you for putting in so much effort-it really shows! Could you please do a video on accessing a RAG chatbot from Microsoft Teams? I’d love to see how it’s done! 😊 Looking forward to more awesome content!
@leonvanzylАй бұрын
Thank you for the feedback 🙏
@Francotujk8 ай бұрын
Excellent content! Very helpful and always appreciated
@tdacarneiro6 ай бұрын
Really appreciate your videos. Greetings from Brazil!
@pioggiadifuoco75228 ай бұрын
Great job Leon, kep going on this, thanks
@leonvanzyl8 ай бұрын
Will do
@CalculatedShots8 ай бұрын
Wish I could sub twice. KZbin does allow 3 watches per user though so Ill end up watching this 3 times or more. We need to help you grow. You deserve it
@leonvanzyl8 ай бұрын
Aww, thanks! Appreciate you ❤️
@DarkKnight-uk7mq8 ай бұрын
Another one. Well done
@RolandoLopezNieto8 ай бұрын
Your videos and information provided are *greatly* appreciated, thanks.
@leonvanzyl8 ай бұрын
You're welcome
@justinchang35738 ай бұрын
Thats crazy i just made a custom chatbot that toggles between 2 chatflows and inserts a mysql query to render cards. What a well times video
@raziiabraham63353 ай бұрын
THIS IS GOLD!!! THANK YOU!!!
@leonvanzyl3 ай бұрын
You're welcome 🤗
@WellPiresAI8 ай бұрын
Very well explained video as always 😊 Waiting for the vid talking about Crew AI!
@leonvanzyl8 ай бұрын
I actually have a video on CrewAI . kzbin.info/www/bejne/bH_Fe4SpeMxkjLM
@gmcwhinney8 ай бұрын
Gteat video as always Leon. Learning a lot from you so I've happily liked & sub'd.
@leonvanzyl8 ай бұрын
Welcome and thank you for the sub!
@hosst8 ай бұрын
Fantastic video Leon, thank you. Do you think this can be used to chain multiple Flowise chatflows?
@leonvanzyl8 ай бұрын
You can use the API tool to call another chatflow using its API Endpoint.
@meister48318 ай бұрын
Hi Leon, this is such great stuff many thanks!
@leonvanzyl8 ай бұрын
You're welcome
@matt_75226 ай бұрын
This video is definetly one of the best videos you ever made! Is it possible to clear the sessions using the API?
@ddd1million8 ай бұрын
Hello! Thank you for the great content! Is there a way to submit an URL via Flowise API, for the "Cheerio Scraper" or "website content crawler" inside a flowise flow to automatically scrape a given URL via API? Basically I have a website with a URL input field, where a person can input a URL and that should be send to the Website crawler inside the flowise flow. Alternatively this would have to be implemented as an Agent with a Web scraper as a tool, if I'm correct? Any advice would be much appreciated, thanks 🙂
@ShuaibShahzan4 ай бұрын
Thanks again Leon for amazing content. Could you please make a detailed tutorial on how to integrate Flowise with WhatsApp and other social media platforms please.
@leonvanzyl4 ай бұрын
Will do. Working on the WhatsApp video.
@ShuaibShahzan4 ай бұрын
@@leonvanzyl Thank you so much :)
@myworld3422 ай бұрын
Thanks Leon! This is a very useful video! What would be the best way to plug VectorShift as the knowledge base and use Flowise as the front end chatbot? I saw Qdrant has a module in Flowise but VectorShift does not. What is the best solution to integrate Vectorshift into Flowise?
@alejandroraulleivailabaca57497 ай бұрын
Thank you for this video @leonvanzyl One question: is it possible to get the sessionId value during an agent runtime? What I'm trying to do is to make an agent use a POST request tool with the sessionId parameter, so I'll have context on which user triggered the request
@josephperkins87668 ай бұрын
Keep them coming 😊
@manuelgarciarubio39088 ай бұрын
Awesome video 😊 I am trying to figure out how to get the tokens used by the conversation retrieval QA chain via the api but i can‘t find a way. I tried to do it with an agent flow and a custom tool…
@leonvanzyl8 ай бұрын
I don't think the API returns the tokens. I'll ask the Flowise team about this. Would be a valuable feature.
@manuelgarciarubio39088 ай бұрын
@@leonvanzyl i think I found a workaround via the langfuse api but if flowise would return that values it could be way easier😅
@fernandogg1528 ай бұрын
Thank u Leon You are greater everyday Luv u❤
@leonvanzyl8 ай бұрын
Thank you! ❤️
@AlexVamp4 ай бұрын
Thank you very much. Your videos are very helpful
@leonvanzyl4 ай бұрын
Glad you like them!
@patousarg8 ай бұрын
Great great video man!!
@leonvanzyl8 ай бұрын
Thank you!
@batguyz7 ай бұрын
One thing I found challenging wads listing Namespaces, there's no clear way to do it, not even in Pinecone docs. Is it also possible to delete upsert data with the API? Great content as always, you are a true champion of Flowise and it's usage!
@junaidmalik6605 ай бұрын
Hi thanks a lot for all the tutorials these are master pieces, I have a question, I am using flowise and a open AI assistant and code interpreter as tool type, now I want to attach a file with code interpreter programmatically, is it possible with any flowise API? or any suggestions? i am able to upload file to open AI using their api, but attaching to coded interpreter is challenging for now
@BrianSoegaard2 ай бұрын
Brilliant work with this video, Leon! 👏 Can you elaborate on how to enable Authorization for the API of a chatbot I want to embed on my homepage? When I enable Authorization as described in the video, my embedded chatbot stops working. It makes no sense to block the (read-only) UI just because I don't want to expose my API, I think. Am I doing it wrong?
@leonvanzyl2 ай бұрын
You're right, the API and chat embed are completely separate. Unless this is a bug, the chat embed should still be working. I'll bring this to the Flowise teams' attention.
@tecnopadre8 ай бұрын
@leonvanzyl it would be great if you can put together a video with the perfect and faster RAG: Reranking, Cohere, etc... Reranking is the best RAGA results before Knowledge Graph which as far as I know can't be implemented on flowise. Tahnk you for the great job you do.
@jeffb63058 ай бұрын
This seems like a good approach to begin monetizing chatbots? Perhaps you can do a video on ways to place chatbots behind a paywall or how to charge users for usage. Thx for all you’re doing !
@leonvanzyl8 ай бұрын
EXACTLY! I've added the API to apps that have user authentication, etc. This is a great idea for a project tutorial. It does involve coding though 😄
@ramsesvandensteen63862 ай бұрын
Stumbling across your channel feels like hitting a goldmine.Your average Joe could spam all your videos and all of a sudden be up there with what engineers at top firms are doing in terms LLM developments
@leonvanzyl2 ай бұрын
This is a massive compliment! Thank you 🙏.
@RenatYaraev8 ай бұрын
Great video, it's true that it's a complicated process so far and it's not much automated. Very interested to know if it will be possible to make a document uploader in user chat in a web application ? Thank you for your hard work!
@leonvanzyl8 ай бұрын
Thank you.
@ritaverissimo17207 ай бұрын
Hi, it seems that with the API endpoints we can customize a lot of things. Is then possible to tweak the AI agent so it's personalized to the user information that the user gave when he signed up to have access? Meaning that the AI tool would greet him by name and have access to all his information (company, preferences, tone of conversation, etc.). Thx!
@martinsmuts25573 ай бұрын
Dankie Leon. Do you know if you can add Flowise to Open WebUI?
@Obinna-ai8 ай бұрын
Great video, very descriptive. Good job, man! Just one small issue - I used the API to upload a PDF vai postman, but when I query it, I get the error "PDFDocument: stream must have data". Am I doing something wrong?
@dxb5268 ай бұрын
Is that even possible via postman? If you're using OpenAI platform the best way (to me) is to create the assistant via Flowise and config/upload everything via OpenAI directly. Otherwise you may run into issues. I had an issue where post request was too large for Flowise, no issue with direct via OpenAI though. Once you're done just sync on Flowise and all changes made on OpenAI are good.
@PeterKuhne-kw9yk8 ай бұрын
I have the same issue!
@MH-xx6dfАй бұрын
How can I get a flowise RAG chatbot on par with a Chatbase one please? There are marked differences with the latter of course being exceedingly advanced.
@ZABBARRIZWI6 ай бұрын
Can you please tell, how can we pass namespace in postgre if we simple wants to do query from a particular namespace.? Here i got answers from every namespace that has been created in the table. But i want it to response from the namespace only which i pass.
@AssassinUK2 ай бұрын
This is great but I have 1 question, most RAG vids miss this off. How do I update the Vector source/Pinecone with new data without duplication. I need to remove/overwrite the previous data?
@leonvanzyl2 ай бұрын
Already have you covered 😁. kzbin.info/www/bejne/qX_OZ2B_iM5qn5Y
@abedmalak_8 ай бұрын
can you please make a video about ZEP memory store and hosting it + using their NEW AI features * NLP, Intent recognition.. that could be gamechanging!!!!
@Young-v4gАй бұрын
Hey Thanks for your tutorial. Do you do any freelance work? If so i am super interested!
@leonvanzylАй бұрын
Hey there. You can send me a message, details in description.
@powerofzero53708 ай бұрын
Brilliant!!
@PIOT238 ай бұрын
Would love to understand a use case for Flowise with Botpress. I’ve been a botpress fan for a while but moved to Flowise since FW seems to satisfy more technical use cases… But how or why you would integrate flowise w botpress id love to see a video or even just learn through comments!
@leonvanzyl8 ай бұрын
The two tools are very different actually. I've been meaning to create a tutorial on Voiceflow and Botpress for a while now. Need to make it happen 😜
@PIOT238 ай бұрын
Making videos as educational and quality as yours takes time, I’ll be patient
@MBStudiosArtАй бұрын
Im getting this error in the chat Cannot read properties of undefined (reading 'incrementCounter')
@BendaDaniel8 ай бұрын
You show that I can override the setting on the fly using curl. Is this option also available through Embed code? Because if I use direct embed in my html code, can someone inject these api calls and change some setting? For example someone can change model that my chatbot uses so then the calls will be more expensive etc. Am I right or there are some protection against injection?
@leonvanzyl8 ай бұрын
No, these overrides are not available in embed
@arthuraquino83568 ай бұрын
Leon is there any way to create the flowise api key externally through an api call?
@leonvanzyl8 ай бұрын
That wouldn't be very secure 🔐.
@misterpsyop8 ай бұрын
Already here!
@sdcharly8 ай бұрын
Hi Leon, do you think the Session Timeouts parameter of UpstashRedisBackedChatMemory works? I tired different values but the session never expires.. i tried to use overrideConfig to change the sessionId from n8n but failed.. not sure how to do it.. Postman works though.
@leonvanzyl8 ай бұрын
I'll check it out. I believe it worked when I tested it initially. PS. I'm working on an n8n tutorial, especially their Langchain / AI nodes. Do you think it's necessary to use Flowise with n8n?
@sdcharly8 ай бұрын
Thanks for the quick response, Leon. n8n has very limited AI exposure still, even Pinecone Serverless, Cheerio Webscraper etc. is not implemented yet. I still feel until its grows up a bit more Flowise still has its space. I'm sure the wide usage options in n8n is always a plus. Me not a python/JS coder will need some help :(. Great work as usual.. you save a lot of sleepless nights :)
@leonvanzyl8 ай бұрын
@@sdcharly you're right, Flowise is miles ahead in terms of Langchain features.
@sdcharly8 ай бұрын
@@leonvanzyl Truly great work.. i was so surprised to see how fast Groq was implemented. Qudos to Henry and other contributors !!
@sdcharly8 ай бұрын
I went thru the source code and found this, --- const sessionTTL = nodeData.inputs?.sessionTTL as string -- sessonTTL being a number.. is this definition correct? I'm not a coder hence maybe I'm totally wrong.
@eugenbabics36148 ай бұрын
Hi Leon, I have a project in mind and I will have to use AI to make it work. I am new to this, no programming experience and for a week I watched almost all of your videos. I understand and managed to duplicate the easier projects but that is not helping with my project. I couldn`t match any of your examples with what I want to do. Are you open to work on a private project?
@leonvanzyl8 ай бұрын
Hi there, send me a mail with your requirements. Check the description of my videos for details
@tarkancorak48882 ай бұрын
Thanks for this great tutorial. I could use Postman to upload a text file and provide the namespace. Do you know how to. specify that namespace in a prediction API call? I tried this, but this did not work: { "question": "...", "overrideConfig": { "pinecone_0": { "pineconeNamespace": "leon-data" } } }
@talhaanwar29118 ай бұрын
did you try embedding the chatbot on a website. i tried but it did not work, though it work with api
@leonvanzyl8 ай бұрын
Yes, I have a dedicated video on embed that you might appreciate. kzbin.info/www/bejne/fWK9hKuBg8mDfZY
@tiagodealexandria66778 ай бұрын
yo can you help me make a flowise flow that it receives the user and password to a determined site as a input and understands when the user give the right credentials and logs him into the site but in the flowise
@leonvanzyl8 ай бұрын
Hey there. That's not possible at this stage. Authentication is specific to the app that's calling Flowise, so you'll have to perform Auth elsewhere and security call Flowise API using to the API key.
@tiagodealexandria66778 ай бұрын
im sorry i didnt understand completely can u try to explain again? Just to make sure: i want thr chatbot to insert the info given by the user to log into the given website
@tiagodealexandria66778 ай бұрын
oh and thx for the quick answer xD
@jeffb63057 ай бұрын
@@leonvanzylcan you recommend a no code frontend that could provide authentication for each user of a Flowise chatbot. This could make it easy to setup a multi-user AI app with Flowise, correct?
@arthuraquino83568 ай бұрын
Thank you very much
@NileValley-TV8 ай бұрын
It is better to put flowise on a server and use from the cloud?
@leonvanzyl8 ай бұрын
What would you like to use it for? If you simply want a chatbot on your own machine, then local is good. For clients and production apps, the cloud is a must
@NileValley-TV8 ай бұрын
@@leonvanzyl I want to start my Ai agency and start hiring out my Ai workers to clients.
@jonasjfsp8 ай бұрын
How can I create my own nodes? I wanted to embed my API and display only its parameters on the screen. already tried? Has anyone seen this?
@leonvanzyl8 ай бұрын
Custom nodes in Flowise? That's news to me 😊 . Which node did you use to do this? I'm under the impression that you would actually have to write a lot of custom code and make changes to Flowise itself to implement such a solution.
@jonasjfsp8 ай бұрын
@@leonvanzyl the idea is hide some nodes to turn instance in some kind of white label, embedding my own logic and integrations, to deliver access to an advanced user
@DhruvAdhia-w4k28 күн бұрын
how to change from v1 to v2 for api endpoints?
@leonvanzyl28 күн бұрын
I think the safer option is to click on the API button on the chatflow and grab whatever endpoint is there.
@NewUser12345-i8 ай бұрын
Hello, can Crew AI be used in Flowise?
@leonvanzyl8 ай бұрын
Technically, yes. Natively, no.. or not yet. CrewAI could be plugged into a custom tool in Flowise. Flowise offers support for Auto GPT, so it won't surprise me if they add support for CrewAI and AutoGen soon.
@abdlrhman900z8 ай бұрын
how can I integrate flowise to my nextjs application?
@leonvanzyl8 ай бұрын
This might be a good idea for a tutorial
@dxb5268 ай бұрын
Idk about nextjs but with react the embed js code works fine. Try to adapt it to nextjs, shouldn't be too different.
@abedmalak_8 ай бұрын
what if the file was hosted on the web? please can we get that???
@leonvanzyl8 ай бұрын
The architecture is that you'd load the file to your server (whether from a UI or File Storage like S3), perhaps in a cache, and then send it off to the API. Really depends on the tech stack and programming language. Happy to create tutorials for different platforms and implementations 👍. I'll probably create a Next JS or Streamlit tutorial soon.