Seedance 2.5: ByteDance extends generative video to 30 seconds in a single shot
The Seed team pushes single-shot generation to 30 seconds, with multi-modal references (30 images + 10 videos + 10 audios) and timestamp editing.
yesterday 8 11
MiniMax H3 open-source: Chinese lab breaks the price of full-modal
MiniMax releases H3, an open-source full-modal (text, image, audio, video) model at one-third the price of competitors. The AI video market stops being binary: expensive closed-source or weak open-source.
Jul 31, 2026 at 12:50 12 14
Gemini Robotics ER 2: DeepMind bets on video-native, multi-robot orchestration
DeepMind's ER 2 update pushes embodied reasoning toward video understanding and fleet-level coordination - the embodied-foundation race gets its Google-branded milestone.
Jul 30, 2026 at 19:40 13 14
FLUX 3: Black Forest Labs announces a multimodal flow model that surpasses Seedance 2.0 and Gemini
Black Forest Labs releases FLUX 3, its new generation of multimodal flow matching models. The lab claims superior results to Seedance 2.0 and Gemini on internal benchmarks. It remains to be confirmed by third-party evaluations.
Jul 24, 2026 at 21:00 8 8
Ropedia raises $22M: the missing "data layer" for robots to understand the world
The foundation embodied models are not lacking algorithms - they are lacking data. Ropedia bets on this layer, between sensor and model.
Jul 24, 2026 at 09:33 9 9
Black Forest Labs releases FLUX 3: a multimodal flow model that spills over into robotics
BFL claims a state-of-the-art shift against Seedance 2.0, Gemini Omni, and Grok Imagine - and in turn pushes a "video-action" derivative for robotics.
Jul 24, 2026 at 09:33 9 8
"AI didn't make programming easier, it just made it differently difficult" - CACM lands the anti-hype line
A CACM article cuts through the noise: AI hasn't simplified programming, it has shifted the difficulty. The new job: arbitrating the outputs of a model that we only partially understand.
Jul 22, 2026 at 12:41 16 10
Manila denounces a racist China state-media AI video: generative video turns into diplomatic ammunition
The Philippines publicly denounce a Chinese state media AI-generated video as racist. Generative video moves from marketing to diplomacy.
Jul 19, 2026 at 00:37 8 8
Kimi K3 preview: Moonshot floods the timeline before shipping
Benchmark clips are everywhere. The model card, the API, and any reproducible evaluation are not.
Jul 16, 2026 at 13:50 32 8
Ronacher: the tower doesn't fall, and that's exactly the problem
Babel stopped when men stopped understanding each other. With the agents, construction continues after the collapse of shared understanding. No one notices anything.
Jul 14, 2026 at 22:30 14 6