Vidu Q1 Launches Cinematic-Grade Visual Effects and Audio Generation To Take On VFX

Vidu, the generative video startup by ShengShu Technology, is back with a new model it calls Q1. This time, the focus is on improvements to quality and control. But to earmark its ascent to the ranks of being among the top of a growing list of generative video competitors, the company is coming out this time with the receipts. VBench, a generative video evaluation standard, announced its ranking and placed Vidu Q1 in first place, putting it ahead of OpenAI’s Sora and Google Gemini.

With new updates rolling out from ShengShu Technology, the company’s latest Vidu Q1 is arguably its attempt to take on the Visual Effects industry. With Hollywood set in its sights – namely to make visual productions easier by lowering the barrier to entry. In fact, recently the company had announced a collaboration with Aura Productions for a 50-episode sci-fi series of AI-generated anime shorts using Vidu.

Vidu Q1 is levelling up with cinematic-grade visual effects that offer generative video clips of up to 5 seconds in 1080p. Additionally, Vidu Q1 offers more fluidity and consistency than before with animated character movements. But there’s more to this iteration.

While hiring for creating visual effects is often a heavy cost, this means that the barrier to entry for those without a crew and editing team like budding social media personalities or an amateur video editor, the bar to create VFX is now a lot lower. The major feature rolling out with Vidu Q1 is called “First-to-Last Frame.” 

It’s a way to create seamless transitions between clips or two images. And the company claims that even if the images are completely irrelevant to one another, its AI algorithm that uses “semantic understanding” manages to find a highly believable way to connect the two together – and it won’t look random. As with generative platforms, using Vidu Q1’s First-to-Last Frame is as easy as typing out some text commands. In one example, if you upload an image of a door and your favorite superheroes and villains, Vidu Q1 will likely output a video of the door opening to the superheroes duking it out.

“Vidu Q1 marks a pivotal step forward in making video generation smarter, more expressive, and more accessible,” said Yihang Luo, CEO of ShengShu Technology. “This new generative video model brings us closer to our vision of building the next-generation content creation platform – one that empowers anyone to turn imagination into reality with unprecedented ease, creative freedom, and cinematic precision.”

Vidu Q1 is also landing a major update that not only introduces visuals but also, for the first time, audio to the platform. “AI Sound Effects” creates high-resolution background music or sound effects, but in an industry first, high-fidelity 48 kHz. And like with its capacity to offer believable transitions, Vidu Q1 offers an audio experience that’s not going to sound ‘off’ or in an uncanny valley. Rather, the audio is without distortions and can layer several audio tracks for up to ten seconds per track. And as a bonus, the generated audio can match the mood of the video as well and be timed specifically to certain timestamps through text prompts – for instance, “add a whooshing sound at 5-7 seconds.”