LTX-2.5
LTX-2.5's matte adapter takes 3.1 minutes on a 16 GB GPU
OMNI Labs timed Clean Plate at 2.9 minutes and Refine Details at 11.2 minutes, and SLOP TV ran none of these tests.
SLOP TV News Desk
kling
Kling 4.0's API is live at 11 to 24 cents a second
The same model runs from $0.106 a second at 720p on Luma's Ultra plan to $0.237 at 1080p on Plus, against Kling's own $0.157 and $0.235.
runway
Runway Ads comes to every paid plan at no extra fee
Runway made Runway Ads available on every paid plan in an October 27 post, after announcing the tool and its Meta, Google and TikTok loop on September 30.
YouTube
YouTube reports 6.5 billion microdrama views in six months
The company is tying its Shorts seasons and episodes to channel memberships, with a members-only episode revamp promised for mobile in the coming weeks.
oracle
Oracle plans Media Fabric, bundling AI video and streaming
Oracle says the platform, combining Creative Studio, AI Media Models and Video Edge, arrives within 12 months; the media models are in beta.
iOS
App Privacy Report: 542 events, and the number a viral post got wrong
iOS's hidden log counts how often apps touch your mic, camera, photos and contacts, and which servers they call. Two posts read 48 hours of it with Claude Code. The log is real; one headline figure is not.
Simulon
Simulon's mammoth is 3D in your room, finished on a server
A three-dimensional asset placed in a scanned room, then finished by a cloud render. The demo that 180,000 people shared has no video model in it.
Higgsfield
Higgsfield Cinema Studio 4.0 builds 30-second AI scenes
A new Director's Panel sets genre, camera, lens and character emotion before generation, and each generation can now draw on up to 50 references, up from nine.
genspark
Genspark Media Studio ships an agent that edits video
The agent plans and edits, runs Claude Opus 5.5 and Fable 5.1 as its director brain, with GPT-6 Astra verifying the 3D stage, and agrees on the script before rendering.
SpAItial
SpAItial's Echo-3 turns phone scans into metric 3D worlds
The world model keeps real-world dimensions and fills in what the camera missed, exporting Gaussian splats, meshes, normals and depth.
Genie 4
Google's Genie 4 makes photoreal worlds you can ride and hear
DeepMind's world model generates the sound with the scene, takes typed instructions mid-simulation and holds a world together for two minutes, double Genie 3. Trusted testers are already using it to prototype shots.
CapCut
CapCut's video editor is coming to ChatGPT as a plugin
CapCut announced the plugin at its CapCut World event in Los Angeles on October 24, and its demo says an open beta reaches ChatGPT in November.
Apple
Final Cut Pro adds built-in MotionVFX effects and templates
The free November update also links Final Cut Pro with Pixelmator Pro and adds beta AI assistants to Logic Pro, Keynote, Pages, and Numbers.
A Woman Asleep
A Woman Asleep debuts at Antalya as an entirely AI-made film
A Woman Asleep screened in the Perge Hall at the Ataturk Cultural Center on October 26, followed by a question session with director Gizem Avcioglu.
Runway
Runway App Studio builds and hosts apps from a description
Runway's changelog lists App Studio on paid plans, with an Adobe panel and ProRes 4444 export listed in earlier entries.
korea
Korea's AI film boom is outrunning its copyright rules
Generative AI has pulled Korean AI films close to live-action quality, but its copyright, likeness and performer rules remain unsettled.
StreamVibe
StreamVibe tests when a video agent should respond
Google researchers and 15 co-authors posted the StreamVibe benchmark, a streaming model and a memory harness for agents that act on user intent mid-task.
ByteDance
ByteDance folds its Xiaoyunque short-drama app into Jimeng
The CapCut unit keeps Xiaoyunque's standalone entry point, but its operations now report to Jimeng's head of operations and marketing.
ByteDance
ByteDance opened a micro-drama factory, and here is how to use it
The studio lives inside Jimeng, which is China-only today, but the pipeline it automates and the models underneath it are reachable from outside through Dreamina and the Seedance 2.5 API. A step-by-step guide to making a 60-episode vertical drama, from premise to cost sheet.
ByteDance
ByteDance's Dreamina builds a studio for AI short dramas
The Jimeng studio writes scripts from novels up to 100,000 Chinese characters and can redraw a finished drama for another region.