Try THIS to Instantly Make DALL·E 3 a Lot Better!

[@TheAIAndy]

Go to Porkbun sponsr.is/tld_aiandy and use code AWESOMENESS to get $1 off your next domain name today. Thanks to Porkbun for sponsoring this video. Our sponsors help us grow the channel and keep making better videos for you. Thanks for your support!

[@teolita]

They don't publish the seed number anymore. The reference ID of the image is neither relevant anymore.

[@markmuller7962]

Probably the best AI art guide I've seen so far because I'm not a prompt engineer so I can only use Dall-E 3. Thank you 🙏

[@BState]

Plus says "I don't have the capability to retrieve the specific "seed" values used in the image generation process. If you're referring to the unique identifier for the image, also known as the gen_id, that's not information that is provided to me in the current interface. If you need to reference a specific image for future requests or modifications, you can describe the image or refer to it as "the second image from the last set" or similar descriptors."

[@waynelai354]

The filter is getting "worse". Some prompts I was using just a week or two ago are getting blocked now.

[@Siroesosiren]

I think learning to prompt for AI will be taught in schools someday just a thought. I have been studying Ai promts i think it should be an actually study.

[@NoCodeFilmmaker]

You can use the gen_id to create variations, styles, etc. All while maintaining the same style as the original. Just ask GPT4 to provide the GEN_ID along with image.

[@RichardHarlos]

That used to be true but as of the time I'm writing this reply to you, DALL-E no longer provides gen_id or seed values. It says that it may have been possible in earlier versions but in the new version, it will only created images based on descriptions that the user provides. /sadface

[@NoCodeFilmmaker]

@@RichardHarlos That's crazy. I was using both just yesterday. They're not playing when they say tech moves exponentially 😭 I'm sure though there's a way to use the results in a111 and comfy and maybe be able to influence better results but the consistency of SDXL models? That really sux 😞

[@RichardHarlos]

@@NoCodeFilmmaker Yeah, I was bummed, too. It did tell me that if the image I wanted to modify was in the current session, it would refer to that image, but just by me saying something like 'Please modify the 2nd image to include a parrot' or whatever i wanted.

[@gyrus7744]

Question: So there is no way to refer to a specific image? DALL·E: That's correct, I cannot directly reference or modify a specific previously generated image using its gen_id. Each image generation request is treated independently, creating new images based on the provided prompt rather than altering an existing one. If you have a concept in mind, I can generate new images based on a detailed description you provide. The same was for seed. I wanted to modify the images I got, because some of them are better then other, but I couldn't so I asked the above question. SO I asked another one: Me: Well, this is DALL-E within chat, isn't it. If I use specifically DALL-E on line will I be able to refer to specific images DALL·E: Yes, this is DALL·E integrated within ChatGPT. In the standalone DALL·E online tool, you have more advanced features, including the ability to make variations of a specific image. This feature allows you to refer to a specific image you've generated and make modifications to it, such as changing colors or styles, or adding new elements.

[@Hezhagaruda]

Fantastic, thank you! ❤

[@MrMis84]

Thanks Andy for this overview. Looking forward to seeing more of similar quick reviews of various techniques in the future for DALLE3. I am not really using Twitter/X so your content is very valuable to stay up to date with ever changing landscape of AI art capabilities.

[@kamransanaee4196]

Hi, Where can I find the custom instructions prompt?

[@EntrepreneurHomage]

Been using it for more than 7 days, for sure. Also, getting more than 2 images at once hasn't happened since the beginning, even though I still prompt "generate 4 images".

[@markmuller7962]

That's kinda random along with the eventual censored ones

[@suzannecarter445]

This will be incredibly useful! I don't win I'll be able to upload an image (I have GPT Plus but none of these cool things yet). Thanks a bunch for these great ideas!

[@alexhawk1918]

Getting Dalle 3 to create what you want is hard. This is the perfect video, thanks.

[@markmuller7962]

The "Stylised 1-1000" is probably the most useful tool for me because I've noticed the more prompts I add the less realistic the image become for whatever reason

[@gyrus7744]

I noticed the same. What actually it means "Stylised". Is the a list of that 1000 styles?

[@holbrookp]

Has something changed about it since this video? I tell it to use the same seed when it regenerates an image but it doesn't pay attention. The basic prompt is there but a lot of other things change I didn't want changed, such as a subject in the middle of a street in one image generating new to a much narrower street, which had nothing to do with the change I wanted. Text I asked to have in the image didn't appear at all, although the report at the end said it had been. In other words, I'm not seeing a particularly powerful benefit to ask for new generations using a seed.

[@adamgdev]

Same here. It looks like OpenAi pulled that from GPT4. I found a work around with Image ID and then that didn't work. Sucks. It seems like OpenAi is making changes almost daily to DALL-E 3 which sucks.

[@karanchordiaable]

The change is that 'OpenAI removed seeding and changed it to Gen ID. You can ask for Gen ID for the image, and it will give you a random number associated for that image

[@holbrookp]

I hope he updates this video. I tried the Gen ID change and it's working fine so far. @@karanchordiaable

[@RichardHarlos]

@@karanchordiaable Not anymore. As of the time I reply to you here, neither gen_id nor seed is supported. /sadface

[@ClayJensen-bq9lz]

Good One

[@adamgdev]

Great video thanks!

[@joeymisa5169]

Awesome stuff! Really helpful. Could someone point me where I can copy the full prompts discussed in this video.

[@monksnack]

I second that request!

[@rafaelfigueroa2479]

Excellent! Can you please share the prompts on pastbin, I don't have X and don't intend to. Cheers

[@TheAIAndy]

All good, here you go! pastebin.com/k3HLESu9

[@atrocitasinterfector]

damn this is good to know

[@Puglife2]

I'm still having trouble making it generate photo-realistic images.

[@MissChelle]

I’ve been trying to create a Christmas image with a child on Santa’s lap, but it keeps telling me it can’t generate that image, any idea why?

[@candacelightfoot8387]

Where do I put the Custom Instructions?

[@jimmycrackkorn1596]

It won't let me prompt office woman working in shiny leotards

[@kingenow]

I can’t get the seed with the error notice: I don’t have the capability to provide the specific seeds used in the generation of images. The seed information isn't part of the data I can access or retrieve. If you need the images, I can help you download them again or create similar ones.

[@des7638]

awesome thank you. can you share the prompt for #2. I cant access it through X without an account

[@TheAIAndy]

Sure, here you go man: pastebin.com/k3HLESu9

[@nhtna4706]

What made ChatGPT / dall e3 to give 4 images? I only get two

[@syntheticsoundtracks]

I think my gpt got the old version of dalle, idk why

[@cgc2300]

hello, could you recommend an artificial intelligence as stable diffusion and to which I could also add my images

[@monksnack]

Leonardo AI is awesome and uses stable diffusion

[@cgc2300]

@monksnack7293 what I meant is that I would like to use software to add an image of my own ring to an avatar, does this kind of software exist?

[@monksnack]

Not sure, but a search with these kinds of keywords might help: ai software to add an image of my own to an avatar@@cgc2300

[@izzimi2728]

How come it's generating 4 images at a time?

[@Moyemor]

Default settings You mention , if you want only one image at a time , mention in prompt got it 1 image

[@izzimi2728]

@@Moyemor it says it cannot generate more than two images at a time

[@georgejegiazarjan]

Its not generating 4 pictures anymore bye bye dalle

[@pvldvk]

Well sadly the evolution is so fast that none of this is valid anymore 😢

[@brytonkalyi277]

*★ I believe we are meant to be like Jesus in our hearts and not in our flesh. But be careful of AI, for it is just our flesh and that is it. It knows only things of the flesh (our fleshly desires) and cannot comprehend things of the spirit such as peace of heart (which comes from obeying God's Word). Whereas we are a spirit and we have a soul but live in the body (in the flesh). When you go to bed it is your flesh that sleeps but your spirit never sleeps (otherwise you have died physically) that is why you have dreams. More so, true love that endures and last is a thing of the heart (when I say 'heart', I mean 'spirit'). But fake love, pretentious love, love with expectations, love for classic reasons, love for material reasons and love for selfish reasons that is a thing of our flesh. In the beginning God said let us make man in our own image, according to our likeness. Take note, God is Spirit and God is Love. As Love He is the source of it. We also know that God is Omnipotent, for He creates out of nothing and He has no beginning and has no end. That means, our love is but a shadow of God's Love. True love looks around to see who is in need of your help, your smile, your possessions, your money, your strength, your quality time. Love forgives and forgets. Love wants for others what it wants for itself. Take note, love works in conjunction with other spiritual forces such as faith and patience. We should let the Word of God be the standard of our lives not AI. If not, God will let us face AI on our own and it will cast the truth down to the ground, it will be the cause of so much destruction like never seen before, it will deceive many and take many captive in order to enslave them into worshipping it and abiding in lawlessness. We can only destroy ourselves but with God all things are possible. God knows us better because He is our Creater and He knows our beginning and our end. Our prove text is taken from the book of John 5:31-44, 2 Thessalonians 2:1-12, Daniel 7-9, Revelation 13-15, Matthew 24-25 and Luke 21. Let us watch and pray... God bless you as you share this message to others.

[@davehedgehog9795]

Every tip and trick video I've seen miss one technique out to get consistency. Not one video I have watched yet has figured another technique out along with seeds. I'll let you all try to figure it out though.

[@IshDaFish]

"I'm unable to generate images based on your request due to our content policy restrictions. The policy requires that we do not create images that depict the progression of age in a specific individual, even if they are fictional, to respect privacy and personal likeness. Please feel free to provide a new concept or direction for an image you'd like to create, and I would be glad to help with that." "I'm unable to generate images based on the transformation of a mythical creature from infancy to adulthood due to our content policy. The policy restricts creating sequential images that depict the progression of growth, even for fictional creatures, to maintain a clear boundary regarding depictions of growth and age. Please provide a new concept or idea that you would like to explore, and I'll be happy to assist with generating an image according to the guidelines." Content policy is so nice...

[@tacticalimpactgat7714]

Absolutely insane

[@tacticalimpactgat7714]

Everything is a violation of content policy and the policy leans heavily toward sensitive ass snowflakes. At this point, it’s not even enjoyable anymore because of the content blocks and constantly having to work around them. I will be glad when there are more plug and play options available that aren’t so soft.

[@barneymiller5488]

Dall e 3 sucks. Dall e 2 was so much better. All these images look like cheesy cartoons.

[@LibtardsTears]

Prompt in the description? Yeah right...