--- title: FLUX 3 - Real World Models: Towards Multimodal Flow... description: Black Forest Labs has announced FLUX 3, a multimodal foundation model that jointly learns from images, video, and audio within a unified architecture. The... canonical: https://daily.dev/posts/flux-3---real-world-models-towards-multimodal-flow-models-as-the-backbone-of-visual-intelligence--8yetuhmwk twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: FLUX 3 - Real World Models: Towards Multimodal Flow Models as the Backbone of Visual Intelligence. | daily.dev og:description: Black Forest Labs has announced FLUX 3, a multimodal foundation model that jointly learns from images, video, and audio within a unified architecture. The... og:url: https://daily.dev/posts/flux-3---real-world-models-towards-multimodal-flow-models-as-the-backbone-of-visual-intelligence--8yetuhmwk og:image: https://api.daily.dev/og/posts/8yEtUhMwk.png og:image:alt: FLUX 3 - Real World Models: Towards Multimodal Flow Models as the Backbone of Visual Intelligence. og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # FLUX 3 - Real World Models: Towards Multimodal Flow Models as the Backbone of Visual Intelligence. **[Hacker News](https://daily.dev/sources/hn)** · 6 min read · 1 upvotes · 0 comments ## Summary Black Forest Labs has announced FLUX 3, a multimodal foundation model that jointly learns from images, video, and audio within a unified architecture. The model is built on their Self-Flow approach and aims to build a unified representation of the world rather than learning individual modalities in isolation. FLUX 3 Video is now in Early Access, capable of generating up to 20-second videos with native audio from text or image inputs, supporting text-to-video, image-to-video, video-to-video, and keyframe-to-video generation. Preliminary evaluations show FLUX 3 outperforming competitors including Runway Gen-4.5 (77% preference) and Luma Ray 3.2 (93% preference). The model also extends to action prediction for robotics, with FLUX-mimic developed in partnership with mimic robotics and tested at Audi. Upcoming releases include FLUX 3 Image, FLUX 3 Action, and an open-weight FLUX 3 Dev variant. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Community take How the wider developer community reacted, aggregated from 2 discussions and 162 comments across hackernews (as of 2026-07-24). **TL;DR:** HN is sharply divided on the FLUX 3 announcement: a minority find the demo genuinely impressive and are excited about the promised open-weight release, while a larger contingent is skeptical of the marketing, critical of the sparse video examples, and concerned about AI-generated slop flooding social media. Much of the thread devolves into meta-debate about HN's own negativity rather than the model itself. **Sentiment:** 25% positive · 40% mixed · 35% skeptical **The case for** - Some commenters who saw early-access clips report the output is stunning and potentially new SOTA for home use. - The promised open-weight release (FLUX 3 Dev) is welcomed by hobbyists who rely on locally-runnable models. - One commenter noted the demo clips looked like real footage until an obviously impossible scene, suggesting high visual fidelity. - Being a European AI lab is seen as a hiring plus by some potential candidates. **The pushback** - The launch showed almost no examples of realistic human faces sustained for more than a few seconds, and used jump cuts rather than continuous 20-second clips as claimed. - The blog post is suspected of being LLM-generated slop, causing some readers to disengage immediately. - Open-weight releases from Black Forest Labs have historically been significantly inferior to the closed versions, and the 'Dev' distilled variant limits fine-tuning effectiveness. - Several commenters argue that text-to-video tools are primarily being weaponized for electoral manipulation, social-media slop, and disinformation rather than beneficial use. - The term 'World Model' is criticized as marketing inflation with no rigorous meaning in this context. **By community** - hackernews (mixed): HN commenters are split between genuine excitement about open-weight potential and sharp skepticism about sparse demos, marketing language, and the societal harms of AI-generated video content, with a large meta-thread about HN's own negativity drowning out technical discussion. **Hottest debate:** Whether AI-generated image/video tools cause net societal harm (via disinformation and slop) or whether such concerns are overblown luddism, mirroring every prior technology panic. **Open questions** - Will the open-weight FLUX 3 Dev model be competitive in quality with the closed API versions, or will it again lag significantly? - Does the model handle realistic human faces and sustained motion without artifacts, given the demo avoided showing them? - Is anyone training models on haptic/touch data, and could that improve robotic manipulation? - What exactly will be missing from the 'Dev' backbone versus the full commercial model? **Highlights** > - Showed close to zero examples of people. - Frivolous use of the term World Model. - Claims 20 seconds of video, shows only jumpcuts. Coming soon! > — [thisisauserid on hackernews · 3 comments](https://news.ycombinator.com/item?id=49032058) > I'm a person who is making extensive use of current-gen "smart" LLM for coding tasks and building automation tools, doing some of the drudge work of gluing disparate open source things together into new projects. I'm fairly optimistic about that. Particularly when you have a good harness setup and you know enough about the subject matter to understand when a model has gone into some dead-end of reasoning or has built something that's not quite right. At the same time, I can see social media is being flooded with absolutely horrid AI slop images and video. I'm much more pessimistic about the practical beneficial real world use of totally artificial image and video generators. It seems the uses that these things are put to when they get into the hands of millions of people are detrimental to society and not a benefit. > — [walrus01 on hackernews · 1 comments](https://news.ycombinator.com/item?id=49033287) > My criticism is specifically with the use of text-to-image and text-to-video generators that will create fully artificial video and image content. I've seen the uses that these tools are being put to in 2025/2026 for electoral manipulation, social issue manipulation and it leaves me with a very distasteful impression. I actively go out of my way to avoid patronizing businesses that advertise with AI slop generated images now, AI slop restaurant menus, and so forth. Whether you want to interpret a desire for authenticity as some sort of self-righteous attitude is up to you. There's a lot of people that share my opinion, and a lot of them that don't. There is also clearly a lot of money behind pushing AI slop images everywhere. Facebook and the various 'pages' and 'groups' that are near 90% AI slop content are a fine example of that. I'm sure a great many advertising impressions and click-throughs have been served, much revenue has been earned. Great success. > — [walrus01 on hackernews · 1 comments](https://news.ycombinator.com/item?id=49033642) > I thought the clips were real footage until they were dancing in a flooded room. > — [Gecko4072 on hackernews](https://news.ycombinator.com/item?id=49032375) > Personally I'm just bored of AI releases. I kinda dread it because it'll hog the front page of HN for a couple of days and it's just the same comments every time: people showing what it generated, people signalling that they aren't luddites etc. For me this is like announcing a slightly more efficient form of coal-powered steam engine. Great, but it's still running on coal. I'm excited about moving beyond coal to cleaner and more sustainable energy sources. Current "AI", based on machine learning, is just recycling existing human works: books, films, code, forum posts etc. But those works, like the coal, is going to run out. We've decided to stop making new things and just burn the coal that's already been deposited. > — [globular-toast on hackernews](https://news.ycombinator.com/item?id=49033377) **Source threads** - [hackernews](https://news.ycombinator.com/item?id=49022910) · 11 points · 0 comments - [hackernews](https://news.ycombinator.com/item?id=49031796) · 195 points · 162 comments ## Similar posts on daily.dev - [FLUX 3 x mimic: The Next Generation of Video-Action Models](https://daily.dev/posts/flux-3-x-mimic-the-next-generation-of-video-action-models-ixm7y902g) · Hacker News · 0 upvotes · 0 comments - [FLUX 3 Multimodal AI: How Black Forest Labs’ Model Beats Seedance 2.0 and Grok](https://daily.dev/posts/flux-3-multimodal-ai-how-black-forest-labs-model-beats-seedance-2-0-and-grok-twdasyenc) · SitePoint · 0 upvotes · 0 comments --- Tags: [#genai](https://daily.dev/tags/genai), [#multimodal](https://daily.dev/tags/multimodal), [#video-generation](https://daily.dev/tags/video-generation), [#flux](https://daily.dev/tags/flux) [View this post on daily.dev](https://daily.dev/posts/flux-3---real-world-models-towards-multimodal-flow-models-as-the-backbone-of-visual-intelligence--8yetuhmwk) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"FLUX 3 - Real World Models: Towards Multimodal Flow Models as the Backbone of Visual Intelligence.","url":"https://daily.dev/posts/flux-3---real-world-models-towards-multimodal-flow-models-as-the-backbone-of-visual-intelligence--8yetuhmwk","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/flux-3---real-world-models-towards-multimodal-flow-models-as-the-backbone-of-visual-intelligence--8yetuhmwk"},"datePublished":"2026-07-24T07:19:55.963Z","dateModified":"2026-07-24T21:32:58.231Z","description":"Black Forest Labs has announced FLUX 3, a multimodal foundation model that jointly learns from images, video, and audio within a unified architecture. The...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/4b6729f03d39037ad4a2fc82ff0e8088?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/4b6729f03d39037ad4a2fc82ff0e8088?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"Hacker News","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Hacker News","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/hn","url":"https://daily.dev/sources/hn"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/flux-3---real-world-models-towards-multimodal-flow-models-as-the-backbone-of-visual-intelligence--8yetuhmwk","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":1},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"genai,multimodal,video-generation,flux","timeRequired":"PT6M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Hacker News","item":"https://daily.dev/sources/hn"},{"@type":"ListItem","position":3,"name":"FLUX 3 - Real World Models: Towards Multimodal Flow Models as the Backbone of Visual Intelligence."}]} ```