Thanks!

Thank you 😊

+1 for GCP deployment ❤ (I think it would deserve quite a few BuyACoffee’s!)

I would love this as well!

Thank you 😊

Render works great! My instance has been running for over a year now and haven't lost my flows 😎. Remember, you do need a paid plan.

You're welcome 🤗

Thank you

You're welcome 🤗

Depends on the client. Some clients prefer hosting it themselves, others prefer that you deal with it. One piece of advice though. Don't use your own OpenAI API key for clients. I usually get them to create their own key and deal with that side of the costs themselves.

You're welcome

Thanks for the info. What will the droplet cost you after those initial free months, out of curiosity? It's been a while, but I remember them being more expensive than Render. Might be wrong

$ 6 /mo $0.009 /hour 1 GB / 1 CPU 25 GB SSD Disk 1000 GB transfer@@leonvanzyl

I agree. Perhaps a dedicated series on deploying to the different service providers is a must.

If you doing so, uploading with the python script is also interesting ​@@leonvanzyl

yes, deploying on Azure would be great

For sure 👍

yes this

Thank you 😊. I actually have a tutorial on removing the branding on my channel 👍.

@@leonvanzyl sorry I didn't search it. Congratulations again for your hard work

It's the same logs as per local. Try setting an environment variable called DEBUG and set the value as true for even more detailed logs.

Thank you 🙏. That's a really good question, wrt moving old chat history to a vector index. Flowise does not offer a feature for this atm, but in productive Langchain and Llama Index apps I would probably implement such a solution. I'm guessing your app has a unique situation where summarising the chat history wouldn't be sufficient?

@@leonvanzyl And how to use chat summation? Or is it already embedded in the circuit? I'm making a character bot that you can just talk to. It won't be good if the old themes are forgotten.

I don't think there is a limit. My one instance has nearly 40 active flows on it and we haven't seen any performance degradation.

Hey, thank you 😊

You can charge way more than $7 per month for hosting their bot. That depends on you. I'll create a separate video on Embedding.

@@leonvanzylIt would be nice a video tutorial from how to move production to ship the bot to a client using Flowise.

May I ask what the issues are that your experiencing?

@@leonvanzyl Sure!!! connecting maridb with Flowise. Load files from a folder, customize the "designed by Flowise"

@@leonvanzyl It is not possible install the mysql database, load folder files, delete the "powered by flowise item"..

Yes! Simply export the local chatflows and import them into your cloud instance.

The deployment will work exactly the same. You will create a fork of the repo and then make your changes to the forked version. So after rebuilding the project, you will then push your changes to GitHub, which will then update the instance on Render.

Thank you Leon. I am still waiting for a more complex use case of output parsers and fetching data from 3rd party APIs :) hopefully, you will have the time for this kind of videos soon@nzyl

Flowise released Multi Modal support a few hours ago. You can now upload images, which are then converted to text. Maybe that's what you're looking for. Not sure about the text to speech part though

@@leonvanzylthanks for the response. I'm just trying to find a way to make flowise read comics aloud. It seems to be quite challenging though

Absolutely, this is all leading somewhere 😁. Next few videos will focus on API and Web Embedding. One thing I haven't quite figured out is how my audience will react to a coding tutorial that uses Flowise. I get an idea that many people who learn Flowise do not want to touch code, so a Nextjs tutorial might be a put off. On the other hand, as a developer I think I should just YOLO and create the tutorials in any event.

@@leonvanzyl actuly I did watch your js langchain playlist it was great for more understanding the whole idea of flowise work. it was explaind in great way in that videos. Keep the great work brother :)

Hehe, I chose Render for a reason. I'm not an infrastructure guy 😁

You would have to find a way to host Ollama in the cloud as well... which is not really in scope for a Flowise tutorial. I'm sure there are some resources out there for an "Ollama Deployment".

Yikes, I haven't experienced this issue myself. I can only think there must be a typo somewhere. Try changing the username and password in environment variables and restart the server.

Thanks I’ll try will that a try!

So something I realized incase anyone else hits this issue. For the environment variables you can't use a double $ sign. It's some bug. So after removing them doing that, it worked.

What's the error/ reason for failure? The above does not include the error unfortunately.

@@leonvanzyl error: failed to solve: process "/bin/sh -c yarn build" did not complete successfully: exit code: 2 error: exit status 1 logs: drive.google.com/file/d/1ADTwAsXDTRhtfhoAXczqzdhWwsY2VjZt/view?usp=drive_link

@@leonvanzyl logs drive.google.com/file/d/1ADTwAsXDTRhtfhoAXczqzdhWwsY2VjZt/view?usp=drive_link

@@leonvanzyl when #15 251.2 [5/5] Building fresh packages... #17 7.450 flowise-ui:build: your devDependencies to work around this error. This will make this message #17 27.49 flowise-components:build: nodes/memory/MongoDBMemory/MongoDBMemory.ts(121,17): error TS2322: Type '{ messages: { $each: StoredMessage[]; }; }' is not assignable to type 'PushOperator'. #17 27.49 flowise-components:build: nodes/memory/MongoDBMemory/MongoDBMemory.ts(177,21): error TS2322: Type '{ messages: { $each: StoredMessage[]; }; }' is not assignable to type 'PushOperator'. #17 27.49 flowise-components:build: nodes/memory/MongoDBMemory/MongoDBMemory.ts(189,21): error TS2322: Type '{ messages: { $each: StoredMessage[]; }; }' is not assignable to type 'PushOperator'. #17 27.70 flowise-components:build: error Command failed with exit code 2. #17 27.72 flowise-components:build: ERROR: command finished with error: command (/usr/src/packages/packages/components) /tmp/yarn--1709292218340-0.5305414795591354/yarn run build exited (2) #17 27.81 ERROR run failed: command exited (2) #17 28.37 error Command failed with exit code 2. #17 ERROR: process "/bin/sh -c yarn build" did not complete successfully: exit code: 2 27.81 ERROR run failed: command exited (2) 28.37 error Command failed with exit code 2. error: failed to solve: process "/bin/sh -c yarn build" did not complete successfully: exit code: 2 error: exit status 1

@@iNevestenko thank you! I'll pass the log file on to the Flowise team to see if it's a known issue. Just a silly question: did you use Docker as the type and did you set the Node environment in the environment variables?

Thank you very much for the support ❤️🙏. It makes a HUGE difference.