Image, video, music and 3D generation — and the work made with it
Alibaba's official Wan account posted a clip from the eighth episode of its live sessions on September 17. Keith Zhang, the filmmaker behind Soulscape, said the 30-second single shot and an omni-reference that takes five videos surprised him most.
By 김현국
Alibaba's official Wan account posted a clip from the eighth episode of its live sessions on September 17. Keith Zhang, the filmmaker behind Soulscape, said the 30-second single shot and an omni-reference that takes five videos surprised him most.
Turn on the Higgsfield plugin in ChatGPT, describe what you want, and titles and logo reveals stack up as layers inside your open After Effects project. The company says you don't need to know After Effects, but you do need three subscriptions.
No photograph or film survives of the day Burt and Ethelle Shatz first met. More than 70 years into their marriage, with his memory fading, Google DeepMind mapped the couple's present-day mannerisms onto their younger faces to rebuild that day.
Handed a single product photo, GPT-6 Astra built a reference frame in Runway, animated it in Blender and saw the final render through. All the human did was say what they wanted.
The model learned from recordings licensed from Warner Music Group, BMG, and Believe. Rights holders receive royalties each time it is used, and older models are being phased out.
On August 31, Higgsfield launched Genjutsu, a tool that swaps out the people and locations in a video without any reshooting. The company also showed a demo linking a phone to a Blender viewport to carry handheld camera movement straight into the render.
Three days after unveiling Solaris, which renders screens, the company has now released GWM Worlds 2 — a real-time world model that adds sound. Why is a video company shifting its focus toward rendering spaces?
Starting from a single scouting photo, the team compared several directions—including a portal concept where crumbling bricks reveal another world—before settling on a final design.
Nick Tasker's 30-second ad "Let's Give Chicken a Break" went from a single base image straight through to final grading
A walkable 3D city drawn entirely in text, with no Unity or Unreal involved, has picked up more than 200 votes on Hacker News
vLLM-Omni and FastVideo's FastH3 have pushed generation time below the video's own runtime
From ninja action to a beverage ad to a yakuza-style cinematic, Higgsfield shows how to turn Blender 3D blocking into Seedance 2.5 video.
It handles not just generated footage but real camera footage too, converting it to 16-bit EXR, ProRes, or HEVC. Available now for Max plan subscribers and above.
A six-era AI short film made as a wedding gift reportedly drew about 500 million combined views across social media, and the creator walked through the entire production process on YouTube
New alpha.midjourney.com, tested by thousands, gets dozens of bug fixes based on feedback
From new image and video generation models to audiobook character casting and music vocal generation, here's a rundown of August's updates
Enter a goal and the agent automatically produces ad and e-commerce assets — yet the obvious domain name is up for sale
No install required — just open the page, and a webcam or phone tilt lets you play a theremin
A workflow that uses GPT Image 2 to change only the headline while keeping layout and logos intact
Short shot with Seedance, with prompts and assets fully released, entered in $1 million film festival
"S4: The Bob Lazar Story," a hybrid of live action and AI, outranked "Avatar" on Amazon's popularity chart
A 20-year David Brent impersonator raised concerns, but what Synthesia builds is corporate training avatars
"KÖK BÖRÜ," based on a Kazakh folktale, was made with Seedance 2.0 and released with its full prompts
Promise blends live-action with real-time backgrounds using Seedance 2.5, while Hicksfield's revenue jumps 35x as the industry resets
