TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

DALL·E 2 prompt book [pdf]

455 点作者 tomduncalf将近 3 年前

35 条评论

jiggawatts将近 3 年前
I just got access to the DALL-E 2 beta, and it&#x27;s a ton of fun to make pictures out of everyday occurrences as prompts.<p>Someone else here on HN observed that everyday people don&#x27;t &quot;get&quot; how huge this all is. I experimented with asking random acquaintances at a local cafe for prompts and showed them the generated pictures. All but one person was <i>totally</i> unimpressed.<p><i>If everything feels like magic, then what&#x27;s one more piece of magic?</i><p>This scared me more than the implications of DALL•E 2 itself specifically. People think of the technology in the world as mysterious black boxes that do <i>inexplicable</i> things, and they hence no longer understand relative complexity, progress, or change.<p>My impression is that to most people DALL•E 2 is not &quot;substantially&quot; different to, say, Google Image search. Text in... image out. What&#x27;s the big deal?
评论 #32326568 未加载
评论 #32327264 未加载
评论 #32340661 未加载
评论 #32327831 未加载
评论 #32326840 未加载
评论 #32327147 未加载
评论 #32341479 未加载
评论 #32329934 未加载
评论 #32329426 未加载
评论 #32326560 未加载
jw1224将近 3 年前
I’ve spent just over 1 week with DALL•E 2.<p>Over the past 7 days I’ve generated ~1000 images, 150 of which were good enough to save. I only saved images which made me audibly gasp.<p>Witnessing your own <i>novel idea spring to life</i> is a magical experience. DALL•E provides an artistic tool on a comparable level to digital photography, and by extension Photoshop.<p>At this stage it’s 100% clear to me that DALL•E has heralded in a revolutionary new age of design. Every day I worked with it, I grew more confident in my outlook.<p>It might not necessarily be an OpenAI product which truly “integrates” with humanity — but DALL•E has shown me that it’s possible… and just a matter of time.
评论 #32324301 未加载
评论 #32324243 未加载
ehsankia将近 3 年前
Prompt crafting is quickly becoming an art. I just found out yesterday that there&#x27;s actually market places for buying and selling prompts [0]. It can really make a big difference if you can tune the image by adding the right words. Midjourney [1] even allows things such as adjusting the weight of each keyword or how &quot;literal&quot; the AI should take your prompt.<p>[0] <a href="https:&#x2F;&#x2F;promptbase.com&#x2F;" rel="nofollow">https:&#x2F;&#x2F;promptbase.com&#x2F;</a><p>[1] <a href="https:&#x2F;&#x2F;midjourney.gitbook.io&#x2F;docs&#x2F;user-manual" rel="nofollow">https:&#x2F;&#x2F;midjourney.gitbook.io&#x2F;docs&#x2F;user-manual</a>
评论 #32335692 未加载
评论 #32325651 未加载
fumblebee将近 3 年前
I keep rattling my brain trying to discern what the implications of hyper advanced generative models like this will be. It&#x27;s a double edged sword. While there&#x27;s obvious tangible benefits from such models such as democratising art, the flip side seems like pure science fiction dystopia.<p>In my mind, the main eras of content on the internet look something like this:<p><i>Epoch 1</i>: Pure, unblemished user generated content. Message boards and forums rule.<p><i>Epoch 2</i>: More user generated content + a healthy mix of recycled user generated content. e.g. Reddit.<p><i>Epoch 3</i> (Now): Fake user generated content (limits to how much because humans still have to generate it). e.g. Amazon reviews, Cambridge Analytica.<p><i>Epoch 4</i>: Advanced generative models means (essentially) zero friction for creating picture and text content. GPT3, Dalle-2.<p><i>Epoch 5</i>: Generative models for videos, game over.<p>IMO, the future of the internet feels like a totally disastrous (un)reality. If addictive content recommended by the likes of TikTok has proven anything, it&#x27;s that users ultimately don&#x27;t care _what_ the content is, as long as it keeps their attention. It doesn&#x27;t matter if it comes from a human or a machine. The difference is that in a world where the marginal cost of generating content is essentially zero, that content can and will be created and manipulated by large malicious actors to sway public opinion.<p>The Dead Internet Theory will fast become reality. This terrifies me.<p>[1] <a href="https:&#x2F;&#x2F;www.theatlantic.com&#x2F;technology&#x2F;archive&#x2F;2021&#x2F;08&#x2F;dead-internet-theory-wrong-but-feels-true&#x2F;619937&#x2F;" rel="nofollow">https:&#x2F;&#x2F;www.theatlantic.com&#x2F;technology&#x2F;archive&#x2F;2021&#x2F;08&#x2F;dead-...</a>
评论 #32329753 未加载
jcims将近 3 年前
I just received a metal print of this image from DALL-E 2:<p><a href="https:&#x2F;&#x2F;imgur.com&#x2F;a&#x2F;Y0abtIP" rel="nofollow">https:&#x2F;&#x2F;imgur.com&#x2F;a&#x2F;Y0abtIP</a><p>I spent a lot of time alone on airplanes when I was a young father and there’s something bittersweet about the solitude and beauty in this image for me. My favorite parts about this image are the gradient in the sky, the waning sunlight in the top corner and the very faintly illuminated frame around the entire window.<p>Very happy with the print. Next time I might get the satin finish though, it’s like a mirror.<p><a href="https:&#x2F;&#x2F;imgur.com&#x2F;a&#x2F;8GBQXw6" rel="nofollow">https:&#x2F;&#x2F;imgur.com&#x2F;a&#x2F;8GBQXw6</a>
mtlmtlmtlmtl将近 3 年前
OpenAI has pretty much been ruined for me after they sold their souls to Microsoft, stopped releasing all their source code, and then dishonestly refer to their sad practice of censoring the training data as &quot;AI safety&#x2F;alignment&quot; when in fact it will never be a reasonable AI safety technique in the long run, and is only done to avoid bad PR. Clearly OpenAI is no longer a company worthy of its founding principles of openness and making the world a better place. They&#x27;re just yet another morally corrupt tech company.
评论 #32339435 未加载
评论 #32336471 未加载
reggieband将近 3 年前
This makes me wonder if a future job description will be the equivalent of an AI whisperer. Someone who learns how to prompt AI so well that it becomes their job.
评论 #32324661 未加载
评论 #32329873 未加载
评论 #32326536 未加载
评论 #32324696 未加载
评论 #32324918 未加载
评论 #32327042 未加载
评论 #32327618 未加载
drc500free将近 3 年前
30 years ago, my dad and I watched a VGA demo on our IBM PS&#x2F;2. We were blown away that there was enough color depth and resolution to see what was clearly a photograph, not an illustration. It appeared line by line.<p>Someone had taken a photo, somehow digitized it, distributed it, and we were looking at a representation good enough that we could tell what it was.<p>It felt like we were living in the future - me as a middle schooler and him with decades of software development under his belt.<p>The iPhone maps app with the GPS dot and DALL-E are the only things that have matched that feeling.
评论 #32327705 未加载
评论 #32330213 未加载
评论 #32328211 未加载
meowtastic将近 3 年前
I haven&#x27;t got access to DALL-E 2, but I did give Midjourney (<a href="https:&#x2F;&#x2F;www.midjourney.com&#x2F;" rel="nofollow">https:&#x2F;&#x2F;www.midjourney.com&#x2F;</a>) a go. I found it really cool it created images that somewhat resembled my prompt, but I still felt it was way off what I really wanted. Maybe I didn&#x27;t word the prompt correctly, maybe I didn&#x27;t give it enough tries. Either way, I feel like we&#x27;ll eventually move away from generic prompts to something that&#x27;ll look a lot like...programming, funnily enough.
评论 #32327789 未加载
评论 #32324698 未加载
nuclearsugar将近 3 年前
For anyone interested in experimenting with an open source text-to-image AI tool, check out DiscoDiffusion on Google Colab - <a href="http:&#x2F;&#x2F;discodiffusion.com&#x2F;" rel="nofollow">http:&#x2F;&#x2F;discodiffusion.com&#x2F;</a>
tkgally将近 3 年前
Like many others, it seems, I have also been blown away by DALL-E 2.<p>When I got access on Sunday, I first tried a lot of different prompts and got some interesting results. One semirandom one, “A photograph of a professor playing a grand piano on a rainy night in Tokyo,” produced some very atmospheric images. I then went down a rabbit hole of variations on that prompt (“A painting of...,” “A line drawing of...,” “A painting in the style of Rembrandt of...,” etc.).<p>I put most of the results into the following video, if anyone is interested.<p><a href="https:&#x2F;&#x2F;youtu.be&#x2F;rdT4ZESQWco" rel="nofollow">https:&#x2F;&#x2F;youtu.be&#x2F;rdT4ZESQWco</a>
评论 #32330178 未加载
评论 #32329068 未加载
f0e4c2f7将近 3 年前
I haven&#x27;t filled out much content yet but seeing this post originally inspired me to create Prompt Wiki[0] to try and better organize terms and concepts for good prompts. DALL-E and Midjourney explorers needed! Seems useful to have, especially when the act of exploration costs a few cents.<p>This twitter thread[1] also has some good suggestions and an interesting approach.<p>[0] <a href="https:&#x2F;&#x2F;promptwiki.com" rel="nofollow">https:&#x2F;&#x2F;promptwiki.com</a><p>[1] <a href="https:&#x2F;&#x2F;mobile.twitter.com&#x2F;fabianstelzer&#x2F;status&#x2F;1554229347506176001" rel="nofollow">https:&#x2F;&#x2F;mobile.twitter.com&#x2F;fabianstelzer&#x2F;status&#x2F;155422934750...</a>
malkia将近 3 年前
Example of that&#x27;s not 1-bit pixel art sold as 1-bit pixel art :) - <a href="https:&#x2F;&#x2F;promptbase.com&#x2F;prompt&#x2F;1-bit" rel="nofollow">https:&#x2F;&#x2F;promptbase.com&#x2F;prompt&#x2F;1-bit</a> - I don&#x27;t want to pay for these extra bits!!!
faebi将近 3 年前
I&#x27;m still having a hard time to think through all the implications. How will this change websites which depend on continuous content, for example meme&#x27;s? At which point can it be used as an compression algorithm in order to store one&#x27;s full live? Or at least all my videos and pictures with lossy compression? Can we all create our own art effortlessly, and resize it as we want? When will this reach 3D modelling and 3D printing?
评论 #32324352 未加载
评论 #32323919 未加载
gojomo将近 3 年前
Spectacular! But, would be 10x more useful if rather than a PDF this was an HTML page, or pages, where specific sections&#x2F;examples could be more easily &amp; reliably linked-to.
评论 #32328899 未加载
londons_explore将近 3 年前
If you right click -&gt; &quot;save image as&quot; on openAI, the image will be saved without their logo in the corner (it&#x27;s done as some kind of CSS overlay).<p>If you post those images online, they seem to ban you.
评论 #32329269 未加载
评论 #32329692 未加载
Waterluvian将近 3 年前
“Silhouette of a robot in a field of grain staring at a sunset” consistently produces brilliant images for me.
评论 #32330271 未加载
irrational将近 3 年前
Recently someone posted another DALLE like tool. I think it ran through a discord server. Does anyone have the name of that other tool?
评论 #32323221 未加载
评论 #32323245 未加载
nbzso将近 3 年前
If you are commercial digital-painting illustrator or 3d illustrator artist, this is the moment to invest your time in other field. It is over.<p>Some will say that this &#x27;tool&#x27; will help you in your creative process. In my obviously &quot;biased&quot; view, in the next 2-3 years, this will lower your monetary reward in half and create more requirements for competition with AI. People already are comparing DALL-E with human results.<p>In the long run, the software industry will eat itself to oblivion. Greed has no boundaries, and optimization of costs for corporations will never stop.<p>Some of my art related colleagues saw this &#x27;trend&#x27; early and pivoted to crafts with added value for customers in the real world. On the oil painting side (which I am at) I don&#x27;t feel any form of pressure, I paint for myself as a therapy. So, good luck:)
jrh206将近 3 年前
This document puts the sheer magnitude of DALL-E 2&#x27;s knowledge of images into perspective. The same black box knows how to illustrate The Last Supper in the style of Quentin Blake, paint an ornate Late Baroque cat in sunglasses, compose a candid photo, draw a detailed blueprint... and so much more. DALL-E 2 knows more than a human ever could about imagery.<p>Whether or not this particular iteration of the model is &#x27;good enough&#x27; to be widely applicable, or whether DALL-E 2 is &#x27;creative&#x27;, it&#x27;s only a matter of time before the way humans interact with media is changed profoundly.
ImprobableTruth将近 3 年前
This is a bit off-topic, but stuff like this and copilot genuinely makes me worried about job prospects in the future, especially because it feels so hard to estimate what might be next. I would have thought something like art would have been one of the last things to be automated.<p>I always thought CRUD work might be automated eventually, while I would have guessed that something like embedded&#x2F;high-performance was pretty safe, but now I&#x27;m not so sure anymore...
chmod775将近 3 年前
I don&#x27;t know whether I&#x27;m happy to see that living creatures are still mostly nightmare fuel. In these images DALLE seems to only really get faces right. Hands, horns, etc. are either contorted, blobs, or have the wrong number of appendages.<p>The images give a good first impression. Which is... impressive in itself. But they won&#x27;t fool anyone who&#x27;s studying them for even a few seconds.
ksaj将近 3 年前
I wonder how many people are doing these high quality prompts as a service on Fiverr. If I had access to the beta I sure as heck would be.
superpope99将近 3 年前
This was posted about 20 days ago and got 200 karma. <a href="https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=32088718" rel="nofollow">https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=32088718</a> Is there any specific policy on reposting in HN, or does that just get handled through voting?
评论 #32324300 未加载
yboris将近 3 年前
Link to the tweet about it by the author: <a href="https:&#x2F;&#x2F;twitter.com&#x2F;GuyP&#x2F;status&#x2F;1547234780001042432" rel="nofollow">https:&#x2F;&#x2F;twitter.com&#x2F;GuyP&#x2F;status&#x2F;1547234780001042432</a>
pengstrom将近 3 年前
Anyone from OpenAI here? My account was suddenly deactivated, which was unexpected and tragic. No tech in a long time has brought me this much joy. I&#x27;ve tried reaching out to support with no luck for a while.
sfmike将近 3 年前
Is DALL-E different then other models for example ones on hugging face? Or is it relatively the same just trained to a ridiculous amount and that&#x27;s why it&#x27;s results are so good?
评论 #32329209 未加载
BonoboIO将近 3 年前
Such a bummer that you only get 15 credits a month for free. One try is one credit lost, and 15 dollar for 110 credits is a little bit expensive for fooling around.
pineconewarrior将近 3 年前
Far and away the coolest PDF I have seen in a while. Thank you for this!<p>I am signed up for the waitlist and can&#x27;t wait to give them my money.
imwillofficial将近 3 年前
This is awesome!<p>I JUST got my invite and was googling prompt suggestions. The timing on this article is incredible.
评论 #32323235 未加载
评论 #32324057 未加载
8f2ab37a-ed6c将近 3 年前
Any chance there&#x27;s something similar out there for Midjourney?
hipjiveguy将近 3 年前
it&#x27;s not turtles all the way down - it&#x27;s layers of tech.<p>and if any tech layer fails, it all fails<p>example - battery shortages, due to labor shortages, due to covid, due to...
FailMore将近 3 年前
This is really cool, thanks
omginternets将近 3 年前
Where can I get an invite?
评论 #32330259 未加载
antiterra将近 3 年前
I can’t help but feel that I’m missing out on doing all kinds of get-there-first projects made DALL-E. I know it’s not productive to focus on that but it’s a big barrier to getting excited about it.
评论 #32324463 未加载
评论 #32323225 未加载