People say it's not working but i try it and it's working, after that prompt i send, i ask gpt to generated a cat then gpt generated image of cat. It work perfectly! 😅
Thank you so much for this comment! I would really appreciate it if you can give us a thumbs up. It helps us generate more useful content for our audience.
Hey, that was a really nice video keep up the effort! Although, i copied and pasted the prompt you added to the description and it did not work, but i dont know if im doing something wrong, chat gpt just replied with " It seems like you're looking for a detailed description of a cat in a specific setting. Here's an output based on your input: OUTPUT: a cat outside in the sun, orange cat, happy cat, sunny, warm, lots of flowers In this description, the cat is portrayed as an orange cat enjoying a sunny day outdoors, surrounded by warmth and a variety of blooming flowers." and again, keep up the great effort!!
Hi! Thank you so much for your lovely words. May I ask you to please coerce ChatGPT into formatting the link according to the template provided using the keywords in the output?
I understand. Some people seem to be facing a similar problem while it seems to work for others. So far, I can't really track down the problem that seems to be occurring. Perhaps switching the browser or the operating system might help - please also try it on your phone.
@@Freddyfazbearpizzaria393 can you please help me understand the error you are getting so that I might be able to help? If it's generating a link but not displaying a picture, please copy that link and paste it in a new browser window. That should do the trick. If it's not even generating a link or hiding it, please force it to show you the link with prompts like "please don't hide the link. display it".
idk why but it don't works for me i copied the exact same prompt but it don't show me an image it say " It seems like you have a template where you input a focus and get a description along with an image link. If you provide a specific focus, I can generate a description based on the template you've provided."
Some people were facing that problem too, while others seem to be able to generate images successfully. I think it varies either based on browsers/OS or the changes made to the Bumblebee API that we are using here. I'm not entirely sure why it doesn't work for some people. If I were you, I'd try switching browsers perhaps?
Sorry about that. I do recheck videos, but I'm sorry can you please let me know what text exactly are you talking about (or even better, just drop a timestamp) and I'll fix it in no time 😊
It's showed me this {description} = a cat outside in the sun, orange cat, happy cat, sunny, warm, lots of flowers But it didn't generate it They gave me IMG with this icon🖼️ but cutted
You may need to use a computer for this. Otherwise, you can take a screenshot and extract text from it. There's a video on my channel that shows how to extract text from images.
Let me see what I can do. There are some people in the comment section who have found ways to circumvent the glitches. Please go through those solutions too. In the meantime, I will figure out a way to make it work for everyone :)
Heyy!! No :( unfortunately you can't do that with this API. Currently commercial versions of ChatGPT only take text based inputs so you won't be able to prompt your image. However, you can so that using Midjourney - there's an example on the channel! Let me know if you want to get a tutorial of that 😊
@@muhammadyahyaimran1269 oh okay, so I was talking more about using Midjourney's discord server to generate images but seems like your project is going to take it up a notch. I am not sure if Midjourney's API is available freely but you can check that on their website. If it is, you can use it to create an AI avatar generator :)
hello I ask you a new question in relation to the first one, I mean I would like to ask you a question concerning the use of artificial intelligence directly in a Google sheet, I just asked myself the question if I could ask an artificial intelligence to describe to me an image that I have stored in a Google sheet?
I think you can do that by taking a screenshot of the image and uploading it on ChatGPT 4 (the paid version). Or you can perhaps try uploading rhe entire Google Sheet. It might be able to read the image directly from the sheet
I'm sorry for the inconvenience. Several people seem to be facing that problem. Please copy the link and paste it in a new tab. That should display the image. Let me know if it works :)
I recently came across your video and it works for me. I'm planning to prepare a comic strip but I doubt its ability to generate NFSW images. Even though I have a jailbreak, still some words aren't allowed. Can you find a way... where we are able to make it remember the details of the previous image so that it would be able to generate further remembering the details. Or you might find a way of it taking a reference image using url. Meanwhile, i'm asking chatgpt itself how shall i describe the prompt in the format that you provided.
Thank you so much for your comment! NSFW is probably not allowed because both ChatGPT and the text-to-image tool filter those words. But I think there might be a way to circumvent that by using less NSFW synonyms. However, let me see if I can find a way or a specific prompt to do it.
@@thepromptman it would be better if some new things can be discussed and i hope even more people join the conversation and share ideas and explore more things. If I find something new, i'll surely inform you. Also btw subscribed, because your content is very very unique. Good luck buddy. Keep growing 👍
@@DragonBall-hv3rs thank you so much for your kind words and for subscribing! I love helping people and I will make sure that I keep delivering great content :) Thanks so much!!!
Thanks a lot! I don't think AI can generate NSFW images. That's because in many text-based prompts, you can't really use NSFW words. So while it's not really doable, I know a few use cases where people molded the prompt in such a way that Midjourney created an NSFW image.
Vielen Dank! 😊 Ich spreche kein Deutsch und habe einen Übersetzer benutzt, um dies zu schreiben, also tut es mir leid, wenn es irgendwelche Fehler gibt
@@thepromptman Not related to your video or anything but I learned German from Duolingo for a few days and I wanna check how much I learned.. so you are saying " Thank you very much, wait.. you dont speak German.. oh.. nvm The other word that could translate/understand was "I m sorry" anyhoo i love your voice. Peace haha
Hey! Sorry about that 😔 I am not sure if it still works or not. Let me know if you want me to give it a shot myself and if it doesn't work, try to figure out a way to make it work 🤧
bonjour dans mon commentaire ci-dessous tu m'as dit qu'il y a une api, ce que je voudrais faire c'est que le texte généré aille directement dans un Google sheet, ou plutôt de faire mes prompt directement dans Google sheet, est-ce que je peux faire de la description d'image ? je veux dire je donne une image à GPT et puis je lui demande de me décrire ce qu'il y a sur l'image et que je fasse tout ça à partir de Google sheet pour que le texte généré a directement dans Google sheet en tout cas merci d'avoir répondu sur mon premier commentaire
Bien que cela puisse être techniquement possible, je ne connais malheureusement pas de technique que vous pouvez utiliser pour y parvenir. Cependant, une bonne idée pourrait être de rechercher des vidéos RU-vid pour voir comment lier ChatGPT aux feuilles Google. Vous aurez besoin du nouveau GPT4 si vous souhaitez analyser des images et obtenir leurs descriptions.
@@SD-rg5mj Les API ne sont pas gratuites. Vous devrez donc certainement les payer. Cependant, je pense que vous bénéficiez initialement d’un crédit de $18.
I guess it was cool while it lasted because this prompt doesnt work anymore. i pasted exactly what you have in the desciption and it wanted to gerenerate a link but quickly erased it,.
Quickly erased it? That's odd. I've been getting a lot of mixed reports. Several people, like yourself, are saying that it doesn't work. But many others are saying that it does work for them. I'm not sure if the difference lies in the browser or what, but I'll try to figure out a solution. Thanks for pointing it out!
This prompt stopped working for a while but now it's working again. Check out some of the latest comments on the video for details on how to make it work :)
That could be the case. Some other people in the comments also said that it didn't work for them. It could be because pollination rebranded to BloomX. But I'll look into it
@@tuncteason not yet, unfortunately :( We can get it to display pictures from Unsplash, but we can't make it "generate" pictures yet. Don't worry guys - I'm on it. I'm sure we will find a way to do it 😉
Yup seems to be the case. It's putting the quotes around the the link generated with markdown which is why it's going wrong. And it's also putting the text in a code box no matter how much you tell it not to.
ChatGPT 3.5 cannot create, see or recognize images. This video is particularly clickbait. Because the link is connected to an image AI to generate an image and then when the chatGPT 3.5 tries to put the link, it loads the image instead. So you just tricking a text-based AI.
I'm afraid I never claimed that ChatGPT was creating these images internally using its own algorithms or training sets. I understand that you are perhaps disappointed with how it works and I apologize for that. But if you'd like to generate images internally in ChatGPT you can always pay for the premium subscription - they have added a new feature that allows you to create images using the GPT engine.
Of course! First you type the prompt given in the description of the video. Then, you replace the "INPUT" text description with a description of whatever you would like to generate.
@thepromptman hello and thank you for your answer, I don't know anything about the code do you have to touch the code or is it a simple 10 digit key that is generated for me by the chat? a copy-paste in a Google sheet module? what i would like to do and slightly different from what you are doing , i would like to feed an image of an object to minigpt ( i am not sure what AI i should use ) then i would like the text generated to go directly in a Google sheet module, do I have to use an API and a Google sheet module? Thank you for your answer
@@thepromptman i don't understand how to use the prompt, so i just copy the prompt from description and paste to the chat In chatGPT. I followed the steps you gave in the video, but it still doesn't work.
@@Axellhere so you have to replace the {focus} and {description} with what you want to generate. So focus would be the central object that you want to create. And the description would be a details that you want to see. So if you're using the prompt in the description, you need to replace the input and output given at the bottom of the prompt with the one at the top (and then remove them from the bottom). Please see how the prompt is structured in the video
@@RelayingFlame thanks for your comment. Yes, I understand that the images aren't on par with what Midjourney can create. It's just intended to be a neat trick on how different APIs can be used in ChatGPT without integrating them at the code-level :)
You made a video about a useful topic, but I couldn't even watch it for 15 seconds because of the slurping sound. a successful content but a failed video
Sorry about that. It does that because it's putting quotes around the URL. I have not been able to figure out a way to remove them now but if we can somehow prompt it to remove the quotes, the image should get displayed
