Build an AI copywriter with Bubble.io and OpenAI's GPT-3
Вставка
- Опубліковано 29 чер 2024
- Learn how to build a nocode AI copywriter with Bubble.io and OpenAI's GPT-3
0:00 - Tutorial Overview
2:46 - Building the UI in Bubble
12:02 - OpenAI Overview
14:40 - GPT-3 prompt design and settings
19:50 - API setup in Bubble
26:47 - Display GPT-3 response on page
32:09 - Split the response into a list
38:40 - Save copy to the database
Follow me on Twitter for all things no-code design, and product:
/ joshmunsch
Support the cause:
www.buymeacoffee.com/joshmunsch
I don’t think you understand just how amazing this tutorial is. You explained everything SO well. Really amazing tutorial! Subscribed!
Hey Joshua, this video is stellar, which got me thinking...why only one video? You're clearly awesome at creating things and explaining how you do it. I'd love to see you put more content like this into the world - I think many others would agree!
Thank you so much! Couldn't have asked for a more straightforward video!
Thanks dude. I've been living on this page for about 3 days now, really appreciate you putting this up.
Absolutely fantastic video. Couldn't have been more crystal clear and clearly described. Many thanks!
Legend! Created this exact AI copywriter and it worked perfectly. Really appreciate you putting in the time and effort to teach us this.
Thank you. Everything is clear, clear and to the point!!!
Fantastic video man! A+ quality instructions and clarity.
Dude you’re a legend. Thanks for such a good tutorial. I don’t normally subscribe to people so liberally but I just subbed to you after watching only half of the video and will keep watching now. Thanks again.
I never comment this type of videos, but you craked amazingly my expectations! Well done!!
This is amazing. Thank you for this. Hoping to see more tutorials like this in the future. 🔥❤
Thank you Josh, this has really been helpful. Excellent tutorial. This was my first time using bubble.
This is exactly what I was looking for. Thank you!
Great video would love to see some more videos like this.
thanks for this video, i spent 5 hours creating this tonight. and created successfully. first ever app. openAI prompts has few things different now. but its easy to understand. pls make more videos you are awesome. happy new year 2023.
Amazing tutorial! Gr8 thx! can't wait for more!!!
Yep, legendary. Followed along exactly to create something similar for a different use case.
Awesome! Thanks a lot man. I have to learn bubble this is what I've been missing.
Really great tutorial!! Thank you
Awesome tutorial! You just learned me how a whole lot more than chatGPT.
Amazing Amazing Amazing!!!! Loved the tutorial, it was really awesome... Thanx alot🔥❤️
Great video!!!
Great video 🙏 thank you ❤️
Very nice!!
Good stuff bro. Saw you off Reddit
Thanks Anshu 🙏
Very very valuable, this is awesome! Thank you so much for sharing this info with us!
One suggestion, can you turn on your mic volume.
Freaking awesome! Thanks
Very nice 💙
great job
Please Continue.... ❤️❤️❤️ Love From Bangladesh 🇧🇩🇧🇩
More coming! Appreciate the kinds words, Mahe 🙏
A Video on how to build an AI Generator using Flutter flow would be KILLLLLER
Excellent Tutorial. Make one with text completion.
Thank you. I've used this as a basis to create a solution on Bubble. Thank you. This has been the most helpful video so far. Please, could you make one for Makersuite & Bubble?
Love it
Great video! In this scenario, if you were to create multiple different generators for different things, is there a benefit to setting up a new call for each specific new prompt in the way you have over just creating 1 API call with a dynamic field in the API Body called , then have a hidden text field in each generator that stores the prompt with dynamic data from the fields? If there is I'd love to know before I build too deep lol
All of this... for free??? Thank you! Instantly subscribed!
Show man!
This is a great tutorial. If you have time, could you let me know if it’s possible to 1) make the generated text appear in the rich text editor and 2) if it can follow on from the end of your written text?
thanks a lot
duuude, how is this your ONLY video?!! you need to do mooore! enlighten us!
Thanks a lot man, can you explain how to show no records in database for new users?
This is awesome! How do you put a limit on the number of requests a user can pull (similar to how the copywriting tools stop at 7000 tokens).
I know it's a nocode but creating an app on bubble is still a challenge for someone like me. Would it be possible to share bubble app template? Amazing tutorial.. please keep up the work.
Thanks for this tutorial! Such a breath of relief for a non-technical dude. I just have a question/request: would you be able to explain how to fine tune a GPT-3 model in a way that doesn't jump straight to entering commands into a window that is unrecognisable to someone non-technical? 🙏
I think I'm part way there now. I downloaded Anaconda prompt for windows.
Thanks Mihira 🙏 I’m actually digging into this currently. Hoping to share more soon
@@artychokes How'd you do?
@@joshmunsch How'd you do?
@@MoonKidAI for my purpose I found that the model works without fine-tuning when the input is formatted in a specific way, so for now I've not gone further down the rabbit hole, instead I've made it clear to the user how to format the input. Not ideal but works for now.
Very nice video. Thank you. I have an application in mind similar to the one you just created in the video. But instead of a list of texts I would like ChatGPT to create a table, and then I need to store the complete table in the database. Any ideas how to do that? The problem is to process correctly in Bubble the text in "choices - first Item
Awesome tutorial!!! But, i have a questions. How can i make the text return in real time, letter by letter?
get issues: sometime, I got incomplete data in response. i have a question how i can remove scrollbar from repeatingGroup element
I have a solution that works about 95% of the time to resolve the previous query. what are the steps in openai to take the feature live in bubble?
Thanks, man for this great tutorial. I have a question, I want to generate a complete article about 800-1000 with a single click, and openai has token limits that don't let me fine-tune the model for that long output. So what is the way to fine tune gpt3 for that long outputs of about 1000 words?
chat gpt already does this for free. You just have to ask it
This is a great video, thank you. I have some questions though... I am requesting 3 suggestions in open ai. In bubble when i run the workflow I get an entry in front of each suggestion, either a "1. text string..." or a "1) text string..." or a "- text string...". Is there a way to maintain a consistent beginning to each text string, or remove anything in front of the text string? Is this bubble or openai? In the raw data in the API call it shows 1:
2:
in front of each suggestion. I have tried to use find and replace controls but this hasn't seemed to have worked. Any advice?
Sounds like it's coming from gpt-3. You can use a regex to find the filter out any characters before the first letter
I can't figure out how to add a copy to clipboard button for the field that generates the information from OpenAI. Help Please!
Plz make a tutorial on "How user can create folders/playlists and Save/Move Posts to folders." in bubble.
I added the save copy button but it's repeating it too many times for every line of text 😬 is there a way I can make sure it only shows per each body
instead of each line?
I'm getting a Raw response for the API: Status code 401 while trying to initialize the API call. I'm doing everything correctly, even tried changing different browsers.
It always gives me this message: You didn't provide an API key. You need to provide your API key in an Authorization header using Bearer auth (i.e. Authorization: Bearer YOUR_KEY)
how does the user retrieve from the database once it's been saved?
Amazing video! I GOING MAD trying to find a way to set a workflow to use automatically the output of the APIcall as an input to get back to another OpenAI call!!!! I know is strong but I need a double pass through the API. Is there anyone out there can HEEEELP ME?! plz :)
Volume is too quiet, if you normalize the audio and add some compression it will even out the levels
Hi this Elie, the Alien. 😀You made an awesome video, but you didn't make any more? Are you disappeared in space somewhere or what?
Create a App like lensa ai avtar creator
How do I contact you for work like this?
Feel free DM me on Twitter @joshmunsch or email me at munschjosh@gmail.com
Can you make ai tool for me paid task?
what u need?
Please slow down next time and explain the steps more! I had no idea what you were doing half the time. For example, it took me 30 minutes to figure out how to multi select something because I thought I was doing it wrong or it was glitched because you didn’t explain how. I probably watched the whole video a total of 3.5 times because of how much I rewinded
Survival of the fittest , Jerry piece !
“Success is peace of mind, which is a direct result of self-satisfaction in knowing you made the effort to become the best of which you are capable.” -John Wooden-.[
What is wrong with people today? I'm still waiting for the AI that will write my UA-cam comments for me.
I can't believe all the idiotic bot comments on this video...
hi. on 27:02 when i go to workflow i clicked to "click here to add an action" like you do. But when i check the Plugins, it shows blank. Please help me? Thank you. @KNOCODE
You just need to change the API from "data" to "action" as he does in 20:39
This is so awesome. Have learnt alot. What if i want to have for example a project on github that has to do with voice tts and work with bubble. How can we achieve this? Now the backend work is all done by openai. hope you can do a video on this.