Awesome -- glad they're going to release the open weight version! I've been waiting for an excuse to re jump into AI OS image gen!

> It jointly learns from images, videos, and audio within a unified architecture, because what it needs to learn is not any one of these elements in isolation.

I'm confused, videos contain images and audio ...?

Amazing. This will be fundamental for the future for robots to distinguish the sound of the poors getting close to Besos/Musk's/Zuckerberg bunkers and quickly adapt to any new kind of attack by the masses, robots will quickly learn to adapt to the behavior of the attackers, quickly infer where they are grouped, their numbers and so for.

Of course there will be feuds from robots of different family groups but they will be minimal as it quickly becomes symmetrical robot conflict with high casualties as they learn too fast from each other, it's likely those will be avoided, it will be after all much easier to confront humans for any given resources.

Truly a pinnacle for technology, albeit perhaps not for mankind.

I wonder if this also creates people with huge heads and short necks like Flux Klein does.

I hope flux will include a 3D generation model. right now the open-weights version of 3D is failing behind closed source by a big margin. Hopefully the improved spatial ability helps with robotics too

i don't get why they are investing money on image/video gen. All generations i have see looked blurry, lacking in fine details and missing the artistic touch (lacks meaning? lifeless?)

Open-weight plans are near the bottom (Launch section):

    - Video and audio generation and editing through APIs and private weight access. (“FLUX 3 Video”)
    - Action prediction through selected research and commercial partners, beginning with mimic robotics (“FLUX-mimic and FLUX 3 Action”)
    - Image synthesis and editing through APIs and private weight access. (“FLUX 3 Image”)
    - Open-weight access to a multimodal backbone, for content creation (video, audio and image) and action prediction. (“FLUX 3 Dev”)

I have a feeling open-weight models ought to be outperforming proprietary ones by now, but that still hasn’t happened. So far, Nano Banana and GPT-2 Image seem to be the best in class, and Flux still isn’t crossing that quality bar.

If you think about it why is language/image even separate from video?

Isn’t video + audio all you need?