Introducing FLUX 3 Image. Control every pixel. Make precise multi-turn edits without changing any other pixel. Lay out the image exactly how you want using bounding boxes. Generate in up to 4K to preserve details. Use up to 10 references to compose an image. Commercial Weights available for companies running image generation at scale. Open Weights version of FLUX 3 Image is launching in the coming weeks.

CLING ON! We’ve got news.

There's an incredibly easy fix for this - LIMITED SLOTS from day one. I can't imagine no one ran the numbers and knew what that number was. That's what being upfront would actually mean.

A former child star hides in a video store. At midnight, his biggest fan finds him. Most people use gen AI for spectacle. The impossible camera moves, the things you can't shoot in real life. I get it. But with this one, I kinda just wanted the film to live or die on faces. Turns out that's the hardest thing I've thrown at Seedance 2. The prompts had to be super dense. Every micro-expression written out, every gesture, every half-second where a face flips from delight to grief to stillness. But there's a limit. If you pack in too much, the model gets overwhelmed and starts hallucinating on every single run. So the whole process was a balancing act. Enough detail to get a real performance, but not so much that the visual falls apart. One character talks a mile a minute, and the other is suppressed, holding everything in. But honestly, the film is for everyone who feels like they peaked a little early. The ones who found some success when they were young, and then watched people take from them, and take, and take, until there was nothing left. Until the thing you loved became the thing that emptied you. It's for the ones who need a little encouragement to get back up. So be nicer to yourself. It's okay to take breaks. Just don't wait twenty years.

Seedance 2.5 is expected to launch around July 4 in the China region via Dreamina AI 🇨🇳✨ Key updates include: • 🎬 Extended video generation length up to 30 seconds • 🔗 Support for up to 50 multimodal references — including image, video, audio, character & style references • ⏱️ Precise temporal control via prompt-based timestamp specification • 🧠 Enhanced Reference-to-Video (R2V) capability, enabling learning from input/output example pairs for improved generation alignment • 🎭 Improved temporal consistency, multi-subject interaction, motion fidelity, and editing-task responsiveness • ⚠️ Reduced IP-generation risk and improved compliance alignment • 🔇 Suppression of unintended subtitles and background music artifacts, with a more natural and less “AI-generated” visual output • 🌍 Expanded multilingual generation capability: supports high-fidelity generation across global languages, including low-resource languages and regional dialects Hope to see a broader international rollout soon 🚀

Excited to announce the films and filmmakers shortlisted for the Next Wave of Cinema Awards @ MetaCannes Film3 Festival. Thx to our Jury, Lauren Neal @dustinhollywood @PAWmckenzie MetaCannes Film3 Festival is an initiative of Film3™, the flagship brand of @Film3Squad

5ives - Ep65 - #aivideo by @VanFalk_ & @kaigani WATCH NOW: youtube.com/watch?v=skZqY3Vf… 5. Play — @DorothyPang_ 4. A Most Elegant Crime — Lion El Aton 3. Shelter — @SCYPHOSinc ⭐ Breakout: The Last Waltz — @WellerVision 2. Spectral — @ButchersBrain 1. Hierarch — @Christian_Luom

The Film Division returns to the Chroma Awards! In Season 1, creators showcased impactful stories across 13 unique categories from Animation to Horror to Comedy. Check out our previous winners at ChromaAwards.com and get ready for our next chapter.

Martin Scorsese is an advisor to Black Forest Labs. He's spent six decades shaping how the world sees stories. Now he's helping us shape visual intelligence with human taste and craft at the center. We sat down with him for a working storyboarding session using FLUX.

They saw it before anyone else Full story dropping tomorrow Stay tuned