Hands down Wes is delivering more value to the community than so many in the space copying off each other. Keep goin bro you should start a Patreon soon so we can support you.
Wow, absolutely LOVE this comment. Thank you so much. I'm trying not to watch any other content to keep my ideas unique and purely me. Glad you're noticing that 😊
0:00: ⚙ A free DALL-E 3 image generator tool was created with an API key, addressing the need for an alternative to existing options. 2:37: 🔑 The video explains how to generate an API key for the openAI platform and walks through the process of setting it up for image generation. 5:11: 📚 The video discusses the new features of the DALL-E 3 model and how it automatically rewrites prompts for safety and detail. 7:43: ⚙ The video explains the process of creating an interface with input boxes, drop downs, and a generate button, as well as the workflows involved. 10:43: 💻 The video discusses a tool with a simple UI interface for generating images using chat GPT, with security measures in place.
🎯 Key Takeaways for quick navigation: 00:00 🆓 *Introduction to DALL-E 3 Image Generator* - Introduction to a free DALL-E 3 image generator using your own API key. - Mention of the motivation behind creating the tool and a promise to cover its usage, features, benefits, backend construction, and potential cost savings. 01:22 🖼️ *Tool Features Overview* - Overview of the features of the DALL-E 3 image generator tool: *- Input API key at the top.* *- Choose image size (standard or HD) and three available options (square, vertical, horizontal).* *- Revised prompt feature to toggle prompt revision on or off.* - Emphasis on the tool's similarity to the OpenAI playground but specifically designed for DALL-E 3 image generation. 03:35 ⚠️ *Data Handling and Tool Disclaimer* - Clarification on data handling: the tool is client-side, and no data (API keys, prompts, or image generations) is saved in the database. - Disclaimer on user responsibility: Users need to save their image generations as nothing is stored on the server. 05:09 🔧 *Revised Prompt Feature Explanation* - Explanation of the revised prompt feature derived from the DALL-E 3 API documentation. - Description of how the tool gives users control over the revised prompt toggle to experiment with obtaining desired image outputs. 06:44 🔄 *Image Generation Walkthrough* - Walkthrough of image generation using the tool, including a demonstration with a prompt for "Spider-Man in a fat suit." - Explanation of the image prompt, loading process, and the resulting HD image. - Highlight of the user's ability to regenerate images and the need to save them locally. 08:57 💲 *Pricing Considerations* - Overview of DALL-E 3 API pricing for different image sizes and qualities. - Comparison with potential savings using the tool versus other methods, such as ChatGPT or GPT-4 Turbo in the OpenAI playground. - Consideration of cost-effectiveness based on the user's specific needs and usage patterns. 10:57 🛠️ *Tool Security and Future Developments* - Assurance of tool security: HTTPS, client-side facing, and no storage of sensitive information. - Suggestions for additional security measures, such as usage limits on API keys. - Tease of potential future developments, including more custom GPTs on the website, based on user interest and feedback. 12:21 🌐 *Accessing the Tool and Closing Remarks* - Invitation to access the tool with a link in the description. - Note about possible URL changes and a commitment to provide updated links in the video description. - Expression of gratitude, ending with another image generation (Hulk as a skinny superhero). Made with HARPA AI
Great work again. It would be really good if you can create something that lets us control the readability level of the articles. I find that most Chat GPT (and others) articles have readability in the Grade 10 to Grade 14 range. For EEAT purposes, I think we need to have it in the Grade 5 to Grade 8 level.
I've had great success with high quality content. I discovered about 6 months ago if you use the instruction in your prompt: use simple English It will definitely improve the overall content plus it removes most of the identifiable AI terms. Good luck
Well done! You are always very clear and practical. I don't have a MidJourney subscription, but I believe it is $10/month for basic and gives you enough compute for roughly 200 images a month, though some have calculated it as near 400 images a month. Just something to think about for people who want image generation.
Thank you for the suggestion! Midjourney is a great alternative, and so is Stable Diffusion (which might be even CHEAPER). I'll be making a few Stable Diffusion XL videos coming up 😊
Thanks for this and a suggested enhancement is to ask GPT to give the user the Gen ID so that they can then use that in the next prompt to alter the image created. I use it all the time.
Great tutorial and tool, I unsubbed from GPT earlier this month and it actually runs out tomorrow. Great to know I can save money and also bypass some of the insane restrictions in GPT4 when creating images.
Yep! I mention this near the end of the video. The major difference with this tool would be the "revised prompt" feature. Bing is going to restrict your generations a bit more.
Hi Wes. Thank you for making this tool available ... Did you find a way to allow use of image seed ID or generation ID? I really miss the very first versions of Dall-E where the user had control over the seed ...
Thank you for this video, what about If I use the wordpress website and I want to add this tool to my website, there is any html code we can have to create this tool.
Hello, Unfortunately, it doesn’t work anymore. There is a 401 error. The message says the api key is incorrect, but it’s not using the personal key entered in the field. The message is exactly the same, whatever API key I entered. Would it be possible for you to fix it, please ? Thanks
@@WesGPT I have the same problem. No matter what you enter in the API key field, you always get: "incorrect API key provided: sk-lgIz2***************************************BcJL." This is also displayed if what you have entered in the field does not even start with "sk-lgIz2". As I also use the API key elsewhere, I am sure that it is correct. I have already cleared the cache and tried several browsers. It would be awesome if you could fix this. Your tool is really great and I use it regularly.
@@dennismulti9459I'm getting the exact same error for the same wrong API key. lgIz2***************************************BcJL. I have 2 OAI API keys, and I get the error returned on both. Neither are anywhere near the one in the message.
You should make a custom GPT that will allow users to make similar images with modifications using the image seed ID or generation ID. Or make one that can let users extract the seed ID from images they created but forgot to get it at the time it was created.
Nice video bro. I have a doubt, I had $5.00 for openai usage limit for for free subscription in openai api key but I am unable to generate image in dall-e-3 model.I am getting error like "Error generating image: RateLimitError: 429 Rate limit exceeded for images per minute in organization". May I know what is the reason. Do I want to increase limit more than $5.00 to generate the image?
Thanks for the walk through. It’s sharpening my brain to look more into bubble 😅. Im really interested in how to download dalle images in csv as png instead of the // format from your article generator app. Is that an easy tweak to your workflow.? Also whats best way to add a request to backend workflow to get html output for Wordpress wpallimport? 🙏
Hmm, good questions. 1. I'll have to look into it but there might be a plugin that can download images (via a URL) to your Bubble Database. Then, when exporting as a CSV, it should download the png image into a cell. Not 100% sure about this though. Is there a reason why you want to do this? 2. I just added this as a feature to the template in my newest video! Check it out here: th-cam.com/video/rA2GlmBtCdM/w-d-xo.html
Hi Wes, Very nice. I'd like to make somting similar in a wordpress website. I do not like to be locked in by a third party like Bubble. Can you show how to proceed with only the free wordpress version in e.g. a small computer (raspberry pi 4/5) with working examples in a github? Many thanks.
Hey, thank you so much for your comment! Sadly, your suggestion is a little bit out of my scope (I'm only somewhat familiar with WordPress and not at all familiar with Raspberry pi 4/5). Sorry about that 😔
Hello. I am a model IRL and I would like to know if there is any AI in which I can load my own image and put myself in different scenarios or generate images of myself in different situations, for example, a medieval vampire or something like that. Thank you and keep up your excellent work.
I believe wes made a video about becoming an AI celeb. If you watch that video, it will give you some insight on how to generate any image with a similar face to the face uploaded (in this case, your face).
Hey, yeah you could generate images with Midjourney and then use the FaceSwap tool! AI Model introduction: th-cam.com/video/p-1YLcdOPIc/w-d-xo.html FaceSwap app tutorial: th-cam.com/video/D1D4_9aqc2Y/w-d-xo.html
The app I made is free to use (if you don't want to pay for ChatGPT anymore and instead save on costs). Also, you might get better outputs directly from the API.
You make great videos on ai generation. Can you please make a video that shows how can one generate various images of an ai generates image in different clothing and at different places but the looks of the person remains the same. Thanks
🎉Where have you been all my life dude? 🎉. Seriously. You have made skills and knowledge and an actual heart 🙏✨🌟 I used Leonardo to generate Spider-Man in a fat suit. 😂😂😂 nsfw
Hands down Wes is delivering more value to the community than so many in the space copying off each other. Keep goin bro you should start a Patreon soon so we can support you.
Wow, absolutely LOVE this comment. Thank you so much.
I'm trying not to watch any other content to keep my ideas unique and purely me. Glad you're noticing that 😊
0:00: ⚙ A free DALL-E 3 image generator tool was created with an API key, addressing the need for an alternative to existing options.
2:37: 🔑 The video explains how to generate an API key for the openAI platform and walks through the process of setting it up for image generation.
5:11: 📚 The video discusses the new features of the DALL-E 3 model and how it automatically rewrites prompts for safety and detail.
7:43: ⚙ The video explains the process of creating an interface with input boxes, drop downs, and a generate button, as well as the workflows involved.
10:43: 💻 The video discusses a tool with a simple UI interface for generating images using chat GPT, with security measures in place.
🎯 Key Takeaways for quick navigation:
00:00 🆓 *Introduction to DALL-E 3 Image Generator*
- Introduction to a free DALL-E 3 image generator using your own API key.
- Mention of the motivation behind creating the tool and a promise to cover its usage, features, benefits, backend construction, and potential cost savings.
01:22 🖼️ *Tool Features Overview*
- Overview of the features of the DALL-E 3 image generator tool:
*- Input API key at the top.*
*- Choose image size (standard or HD) and three available options (square, vertical, horizontal).*
*- Revised prompt feature to toggle prompt revision on or off.*
- Emphasis on the tool's similarity to the OpenAI playground but specifically designed for DALL-E 3 image generation.
03:35 ⚠️ *Data Handling and Tool Disclaimer*
- Clarification on data handling: the tool is client-side, and no data (API keys, prompts, or image generations) is saved in the database.
- Disclaimer on user responsibility: Users need to save their image generations as nothing is stored on the server.
05:09 🔧 *Revised Prompt Feature Explanation*
- Explanation of the revised prompt feature derived from the DALL-E 3 API documentation.
- Description of how the tool gives users control over the revised prompt toggle to experiment with obtaining desired image outputs.
06:44 🔄 *Image Generation Walkthrough*
- Walkthrough of image generation using the tool, including a demonstration with a prompt for "Spider-Man in a fat suit."
- Explanation of the image prompt, loading process, and the resulting HD image.
- Highlight of the user's ability to regenerate images and the need to save them locally.
08:57 💲 *Pricing Considerations*
- Overview of DALL-E 3 API pricing for different image sizes and qualities.
- Comparison with potential savings using the tool versus other methods, such as ChatGPT or GPT-4 Turbo in the OpenAI playground.
- Consideration of cost-effectiveness based on the user's specific needs and usage patterns.
10:57 🛠️ *Tool Security and Future Developments*
- Assurance of tool security: HTTPS, client-side facing, and no storage of sensitive information.
- Suggestions for additional security measures, such as usage limits on API keys.
- Tease of potential future developments, including more custom GPTs on the website, based on user interest and feedback.
12:21 🌐 *Accessing the Tool and Closing Remarks*
- Invitation to access the tool with a link in the description.
- Note about possible URL changes and a commitment to provide updated links in the video description.
- Expression of gratitude, ending with another image generation (Hulk as a skinny superhero).
Made with HARPA AI
OMG, thank you so much for creating the custom GPT to enable HD mode. I've been trying every which way to get DE3 to force HD mode, and now I can! :)
Oh so awesome that this tool helped you out!
Great work again. It would be really good if you can create something that lets us control the readability level of the articles. I find that most Chat GPT (and others) articles have readability in the Grade 10 to Grade 14 range. For EEAT purposes, I think we need to have it in the Grade 5 to Grade 8 level.
I believe Mr. Beast scripts his videos at a 3rd grade level so having 5th to 8th grade would be pretty nice.
This is a great suggestion! Let me see what I can do and get back to you.
Is there a website that you use to grade the readability of text?
@@WesGPTGrammarly has a built in feature for that
I've had great success with high quality content. I discovered about 6 months ago if you use the instruction in your prompt: use simple English
It will definitely improve the overall content plus it removes most of the identifiable AI terms. Good luck
Well done! You are always very clear and practical. I don't have a MidJourney subscription, but I believe it is $10/month for basic and gives you enough compute for roughly 200 images a month, though some have calculated it as near 400 images a month. Just something to think about for people who want image generation.
Thank you for the suggestion! Midjourney is a great alternative, and so is Stable Diffusion (which might be even CHEAPER). I'll be making a few Stable Diffusion XL videos coming up 😊
Thanks for this and a suggested enhancement is to ask GPT to give the user the Gen ID so that they can then use that in the next prompt to alter the image created. I use it all the time.
I haven't had much luck using the Gen ID since the latest update! Have you been able to get consistent images this way?
Doesn't work. Don't even get a loading bar.
Great tutorial and tool, I unsubbed from GPT earlier this month and it actually runs out tomorrow. Great to know I can save money and also bypass some of the insane restrictions in GPT4 when creating images.
Yeah, we're in the same boat here - loving the lower restrictions and cheaper generations by using the API
Could you do the same thing with Bing Image create that uses Dall-E-3? You don't need a key for it.
Yep! I mention this near the end of the video.
The major difference with this tool would be the "revised prompt" feature. Bing is going to restrict your generations a bit more.
Hi Wes. Thank you for making this tool available ... Did you find a way to allow use of image seed ID or generation ID? I really miss the very first versions of Dall-E where the user had control over the seed ...
Thank you for this video, what about If I use the wordpress website and I want to add this tool to my website, there is any html code we can have to create this tool.
This is great... Please keep building in Bubble.
I definitely will 😊
Hello, Unfortunately, it doesn’t work anymore. There is a 401 error. The message says the api key is incorrect, but it’s not using the personal key entered in the field. The message is exactly the same, whatever API key I entered. Would it be possible for you to fix it, please ? Thanks
Just tested and it still works! Make sure your API key is correct and then try again 😊
im with the same problem, can you fix that?@@WesGPT
@@WesGPT I have the same problem. No matter what you enter in the API key field, you always get: "incorrect API key provided: sk-lgIz2***************************************BcJL." This is also displayed if what you have entered in the field does not even start with "sk-lgIz2". As I also use the API key elsewhere, I am sure that it is correct. I have already cleared the cache and tried several browsers. It would be awesome if you could fix this. Your tool is really great and I use it regularly.
@@dennismulti9459I'm getting the exact same error for the same wrong API key. lgIz2***************************************BcJL. I have 2 OAI API keys, and I get the error returned on both. Neither are anywhere near the one in the message.
Can you do a video on how to use assistant gpt's and put that on a website via api?
I'll add that suggestion to the video pipeline!
Nice, the problem for me is a lot of the stuff I do involves uploading an image as a reference point. and this tool doesn't allow for that.
I think I can add that as a feature for you!
@@WesGPT I'll buy you a beer if you do!
@@MexieMex Nah, no need! Let me see what I can do.
You should make a custom GPT that will allow users to make similar images with modifications using the image seed ID or generation ID.
Or make one that can let users extract the seed ID from images they created but forgot to get it at the time it was created.
Ahh, good idea! Let me see if I can add that as a feature 😊
hey man. you think maybe you can give an in-depth tutorial on how to build my own version of this? yours has seemed to reach its limit
Definitely, I think this is a GREAT idea!
Nice video bro. I have a doubt, I had $5.00 for openai usage limit for for free subscription in openai api key but I am unable to generate image in dall-e-3 model.I am getting error like "Error generating image: RateLimitError: 429 Rate limit exceeded for images per minute in organization". May I know what is the reason. Do I want to increase limit more than $5.00 to generate the image?
Nah, this shouldn't be dollar limit related. I haven't seen this error before so try contacting OpenAI's support!
Thanks for the walk through. It’s sharpening my brain to look more into bubble 😅. Im really interested in how to download dalle images in csv as png instead of the // format from your article generator app. Is that an easy tweak to your workflow.? Also whats best way to add a request to backend workflow to get html output for Wordpress wpallimport? 🙏
Hmm, good questions.
1. I'll have to look into it but there might be a plugin that can download images (via a URL) to your Bubble Database. Then, when exporting as a CSV, it should download the png image into a cell. Not 100% sure about this though. Is there a reason why you want to do this?
2. I just added this as a feature to the template in my newest video! Check it out here: th-cam.com/video/rA2GlmBtCdM/w-d-xo.html
Dalee amigo! thank you for the apps. I wonder if you could do something similar using the Gemini API?
So I can't yet! I live in Canada and it's not available to us.
But as soon as I get access, I will 💯
Does not work anymore, says api key is incorrect even though it is
Sorry, try again now!
Hi Wes, Very nice. I'd like to make somting similar in a wordpress website. I do not like to be locked in by a third party like Bubble. Can you show how to proceed with only the free wordpress version in e.g. a small computer (raspberry pi 4/5) with working examples in a github? Many thanks.
Hey, thank you so much for your comment!
Sadly, your suggestion is a little bit out of my scope (I'm only somewhat familiar with WordPress and not at all familiar with Raspberry pi 4/5). Sorry about that 😔
Hello. I am a model IRL and I would like to know if there is any AI in which I can load my own image and put myself in different scenarios or generate images of myself in different situations, for example, a medieval vampire or something like that. Thank you and keep up your excellent work.
I believe wes made a video about becoming an AI celeb. If you watch that video, it will give you some insight on how to generate any image with a similar face to the face uploaded (in this case, your face).
Hey, yeah you could generate images with Midjourney and then use the FaceSwap tool!
AI Model introduction: th-cam.com/video/p-1YLcdOPIc/w-d-xo.html
FaceSwap app tutorial: th-cam.com/video/D1D4_9aqc2Y/w-d-xo.html
Unfortunately doesnt seem to be working
Great work. Learning from your videos
So glad they're helping!
ChatGPT doesn't let me generate spider man for being a copyrighted character.
Try disabling the revised prompt toggle - does it work then?
I’m confused. How is using your API key free?
you still get an API charge, just the interface he made is free.
The app I made is free to use (if you don't want to pay for ChatGPT anymore and instead save on costs). Also, you might get better outputs directly from the API.
hey your site doesn't work now.
Hey, I just tested and still works! Which error message are you getting?
You make great videos on ai generation. Can you please make a video that shows how can one generate various images of an ai generates image in different clothing and at different places but the looks of the person remains the same. Thanks
Let me look into this and get back to you!
ig it doesnt work anymore i cant generate even if i used an api key
It should work! Make sure you have credits in your OpenAI account.
Thank you so much, you have save me lot of time, God bless you.
You're so welcome! God bless you too.
great stuff! Love the channel
Appreciate having you on board 🙏
Perfect. Thank you very much!
You're so welcome!
Does not funtion anymore:
There comes the failure that API KEy is wrong.
Try again now!
This was great woohoo love ur channel
Thanks 😊
It's not working now.
Sorry about that! We're on it 🚀.
Nice video but you did not delete api key that you create in the past video so please delete 😊
Ahh thank you! I've deleted almost all of them now.
Can you build me a tool
Yes, of course! What are you looking for?
super cool brother
Glad to hear 🙏
Awesome as usual
Thank you so much for the support 🙏
👋
Appreciate it 🙏
🎉Where have you been all my life dude? 🎉.
Seriously. You have made skills and knowledge and an actual heart 🙏✨🌟
I used Leonardo to generate Spider-Man in a fat suit. 😂😂😂 nsfw
Dude, I've been right here!
Going to check out Leonardo and make a video about it for the AI model series.
So it's not FREE like your title smh. Dislike
@WesGPT Fantastic, thank you... 🔥🔥🔥🔥!!
Absolutely no problem at all 😊
Dalle3 vs Midjourney v6 🤔
Oh this is a good idea for a video! Prompt battle!
@@WesGPT yep 💯
🔥