TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

FLUX1.1 [pro] – New SotA text-to-image model from Black Forest Labs

228 pointsby fagerhult8 months ago

22 comments

in3d8 months ago
Better link <a href="https:&#x2F;&#x2F;blackforestlabs.ai&#x2F;announcing-flux-1-1-pro-and-the-bfl-api&#x2F;" rel="nofollow">https:&#x2F;&#x2F;blackforestlabs.ai&#x2F;announcing-flux-1-1-pro-and-the-b...</a>
vessenes8 months ago
Flux is so frustrating to me. Really good prompt adherence, strong ability to keep track of multiple parts of a scene, it&#x27;s technically very impressive. However it seems to have had no training on art-art. I can&#x27;t get it to generate even something that looks like Degas, for instance. And, I can&#x27;t even fine tune a painterly art style of any sort into Flux dev. I get that there was working, living artist backlash at SD and I can therefore imagine that the BFL team has decided not to train on art, but, it&#x27;s a real loss. Both in terms of human knowledge of, say composition, emotion, and so on, but also for style diversity.<p>For goodness sake, the MET in New York has a massive trove of open CC0 type licensed art. Dear BFL, please ease up a bit on this, and add some art-art to your models, they will be better as a result.
评论 #41732742 未加载
评论 #41733921 未加载
评论 #41732534 未加载
评论 #41732542 未加载
评论 #41732817 未加载
评论 #41741180 未加载
评论 #41738492 未加载
评论 #41732716 未加载
评论 #41734210 未加载
ilaksh8 months ago
Pretty smart model. Here&#x27;s one I made: <a href="https:&#x2F;&#x2F;replicate.com&#x2F;p&#x2F;6ez0x8xqvsrga0cjadg8m7bah0">https:&#x2F;&#x2F;replicate.com&#x2F;p&#x2F;6ez0x8xqvsrga0cjadg8m7bah0</a>
评论 #41735144 未加载
评论 #41733193 未加载
评论 #41734163 未加载
评论 #41733857 未加载
评论 #41732858 未加载
评论 #41734972 未加载
ChrisArchitect8 months ago
Announcement post: <a href="https:&#x2F;&#x2F;blackforestlabs.ai&#x2F;announcing-flux-1-1-pro-and-the-bfl-api&#x2F;" rel="nofollow">https:&#x2F;&#x2F;blackforestlabs.ai&#x2F;announcing-flux-1-1-pro-and-the-b...</a><p>(<a href="https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=41730626">https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=41730626</a>)
sharkjacobs8 months ago
&quot;state of the art&quot; has become such tired marketing jargon.<p>&quot;our most advanced and efficient model yet&quot;<p>&quot;a significant step forward in our mission to empower creators&quot;<p>I get it, you can&#x27;t sell things if you don&#x27;t market them, and you can&#x27;t make a living making things if you don&#x27;t sell them, but it&#x27;s exhausting.
评论 #41732562 未加载
评论 #41732146 未加载
评论 #41732304 未加载
评论 #41732516 未加载
评论 #41732414 未加载
评论 #41732113 未加载
Der_Einzige8 months ago
Far more interesting will be when pony diffusion V7 launches.<p>No one in the image space wants to admit it, but well over half of your user base wants to generate hardcore NSFW with your models and they mostly don’t care about any other capabilities.
Jackson__8 months ago
Ah, that was one short gravy train even by modern tech company standards. Really wish the space was more competitive and open so it wouldn&#x27;t just be one company at the top locking their models behind APIs.
skybrian8 months ago
It doesn’t get piano keyboards right, but it’s the first image generator I’ve tried that sometimes get “someone playing accordion” mostly right.<p>When I ask for a man playing accordion, it’s usually a somewhat flawed piano accordion, but If I ask for a woman playing accordion, it’s usually a button accordion. I’ve also seen a few that are half-button, half-piano monstrosities.<p>Also, if I ask for “someone playing accordion”, it’s always a woman.
评论 #41732615 未加载
whitehexagon8 months ago
I&#x27;m running Asahi Linux on a 32GB M1 Pro. Any chance of being able to run text-to-image models locally? I&#x27;ve had some success with LLMs, but only the smaller models. No idea where to start with images, everything seems geared towards msft+nvda.
评论 #41734906 未加载
评论 #41734090 未加载
评论 #41734022 未加载
doctorpangloss8 months ago
I&#x27;m worried about what happens when more people find out about Ideogram.<p>There are a lot of things that don&#x27;t appear in ELO scores. For one, they will not reflect that you cannot prompt women&#x27;s faces in Flux. We can only speculate why.
评论 #41732132 未加载
评论 #41732074 未加载
byteknight8 months ago
I won&#x27;t pay for a model, but that cake image looks dang good.
评论 #41740080 未加载
evrim1891118 months ago
I think Flux is better than SDXL and Dall e. I tried the models from here <a href="https:&#x2F;&#x2F;apps.apple.com&#x2F;us&#x2F;app&#x2F;art-x-a-i-art-generator-aiart&#x2F;id1644315225" rel="nofollow">https:&#x2F;&#x2F;apps.apple.com&#x2F;us&#x2F;app&#x2F;art-x-a-i-art-generator-aiart&#x2F;...</a>
fortran778 months ago
I&#x27;ve been playing with Flux.Dev and such a big step forward from Stable Diffusion and all the other Generative AIs that could run on consumer GPUs.<p>I just tried this Flux1.1 pro page (prompt: &quot;A sad Macintosh user who is upset because his computer can&#x27;t play games&quot;) and was very impressed by the detail and &quot;understanding&quot; this model has.
jeffbee8 months ago
I asked for a simple scene and it drew in the exact same AI girl that every text-to-image model wants to draw, same face, same hair, so generic that a Google reverse image search pulls up thousands of the exact same AI girl. No variety of output at all.
ks20488 months ago
Is there a good site that compares text-to-image models - showing a bunch of examples of text w&#x2F; output on each model?
nirav728 months ago
Are there any projects that allow for easy setup and hosting Flux locally? Similar to SD projects like InvokeAI or a1111
评论 #41732556 未加载
评论 #41732399 未加载
评论 #41732495 未加载
评论 #41732462 未加载
评论 #41735102 未加载
评论 #41732091 未加载
评论 #41732228 未加载
评论 #41738572 未加载
评论 #41732686 未加载
kindkang20248 months ago
I really enjoy its service. It&#x27;s promising for UI design. My advocacy website pages&#x27; UI design was bootstrapped using it. It is quite good for developers without much design ability.<p>Ironically, I am afraid to type the website out and will keep it unknown here. My account could be suspended because of this. It had already reached -1 karma. It&#x27;s better to keep my account alive.
nubinetwork8 months ago
I tried using schnell, it won&#x27;t fit in a 16gb GPU, and I couldn&#x27;t get it to run on CPU.
评论 #41740141 未加载
评论 #41738741 未加载
Mashimo8 months ago
Oh neat. I wonder if they also improve .schnell and .dev soon. That would be nice :)
jchw8 months ago
The generated images look impressive of course but I can&#x27;t help but be mildly amused by the fact that the prompt for the second example image insists strongly that the image should say 1.1:<p>&gt; ... photo with the text &quot;FLUX 1.1 [Pro]&quot;, ..., must say &quot;1.1&quot;, ...<p>...And of course, it does not.
评论 #41732165 未加载
ionwake8 months ago
Sorry to be a noob, but how does this relate to fastflux.ai which seems to work great and creates an image in less than a second? Is this a new model on a slower host?
melvinmelih8 months ago
In case you want to try it out without hassling with the API, I&#x27;ve set up a free tool for it so you can try it out on WhatsApp: <a href="https:&#x2F;&#x2F;instatools.ai&#x2F;products&#x2F;fluxprovisions" rel="nofollow">https:&#x2F;&#x2F;instatools.ai&#x2F;products&#x2F;fluxprovisions</a>