ByteDance’s Seedance 2.0 marks new era of cinematic AI-generated videos: Here’s how
Users and analysts are calling Seedance 2.0 'game changer' and 'next evolution of AI video'
ByteDance has introduced a new AI video generation model Seedance 2.o in China, capable of creating up to two-minutes 1080 clips from simple prompts, containing text, images, audio or video.
Since its launch, the AI video generator has taken the internet by storm, with some calling it a “game changer” and “next evolution of AI video.”
Seedance 2.0 creates multi-shot sequences with consistent characters, sound effects, realistic physics-based actions, music, and voiceovers.
As reported by The Information, the latest addition signals China’s growing foothold in the advancements of AI video technology, giving the country an edge in competitive AI landscape.
The other features of Seedance 2.0 include native multi-shot storytelling from a single prompt, Phoneme-level lip-sync in 8+ languages, and 30 percent faster generation than v1 via RayFlow optimization.
The generated videos exhibit 1080p cinematic quality, maintain motion consistency across shots, multi-shot coherence, and audio-video joint generation.
According to the users, the Chinese video generating app can outperform Sora 2, Veo 3.1, and Kling 3.0.
Talking about its availability, only a limited number of users of Jimeng, ByteDance’s AI image and video app in China, and Jianying, the company’s Chinese video editing app can access to Seedance 2.0.
“The actual test results are impressive, supporting input such as text, images, videos, and audio, and has made breakthroughs in key capabilities,” Kaiyuan Securities analysts including Fang Guangzhao said.
Soon after video model launch, Chinese media, AI application firms, and AI app stocks have witnessed a surge in market rally.
Social media platforms, such as X are flooded with cinematic AI videos created by Seedance 2.0.
Some users showed excitement. One user wrote, “What it does (kinda like Sora but more advanced) is act like a director being able to create entire full length videos with lots of cuts and different scenes but with you just giving it one prompt.”
While others raised concerns regarding the potential job losses in editing, scripting, and production.
-
Mark Zuckerberg reassures employees, says 'no further company-wide layoffs' expected this year
-
Is AI boom a bubble? Jeff Bezos has its surprising answer
-
Jeff Bezos says space data centres are 'very realistic', but not soon
-
Bristol Myers to deploy Anthropic’s Claude AI model to accelerate drug discovery
-
US groups demand Roblox investigation over child safety, ‘deceptive’ marketing practices
-
Zuckerberg leaked audio shows Meta used employees to train AI
-
OpenAI co-founder Andrej Karpathy joins Anthropic to lead AI research unit
-
AI layoffs are ‘dumb’, says Google DeepMind chief Demis Hassabis
-
Could AI create $500 trillion global economy? Elon Musk, Jensen Huang think so
-
GitHub probes alleged breach of 4,000 internal repositories
-
Why Singapore is urging financial firms to scale up AI for better jobs
-
Google reinvents search with AI in biggest update in 25 years: Is web traffic apocalypse coming?
