Veo 3.1, the latest version of Google’s popular AI video generation tool, is starting to become available via third-party AI video generation services, and it appears it will be able to support multi-prompting for multi-shot effects, as well as generate videos at 1080p that are up to a minute long.
What made Google’s Veo 3 different from previous AI video generators was its ability to lip-sync audio with characters in the video. In Veo 3, lip-sync, voiceovers, music, and effects were perfectly aligned in a single video. Veo 3 was quickly added to Google’s Gemini Flash 2.5, which is its current model, so Gemini users could create AI video from the chatbot text prompt.
While the latest Veo 3.1 is taking things even further, it isn’t available in Gemini yet. However, you can access it via third party AI video generation services like Higgsfield, Imagine Art and Envato.
You may like
Over 30-second-long clips
The Higgsfield website says: “Whether starting from text-to-video or image-to-video prompts, Veo 3.1 ensures characters stay consistent across every frame, environments transition naturally, and camera angles shift with the precision of a real production studio.”
This sounds much more involved than the simple eight-second clips that Veo 3 produces. The Higgsfield website also lists “30 seconds+” video clips as a benefit of Veo 3.1.
The Imagine Art website says: “With Google Veo 3.1, you don’t have to look twice to ensure character and scene consistency. It captures character and scene interaction with perfection.”
Veo 3.1 also comes with Cinematic Presets. These give you complete control of the narrative of your visual storytelling, so you can easily incorporate complex effects without having to do all the prompting hard work yourself. Presets include drone shots, slow or fast pans, zoom in and out, tracking shots and more. Lighting and tone presets are also available.
Multi-prompts
A social media post by AI film maker Volodymyr Cherner claims that Veo 3.1 will be able to generate one-minute-long video clips.
Listing all the benefits of Veo 3.1, Cherner writes, “Among the key features meant to justify the ‘.1’ in the name are: character consistency. Now, your generated hero won’t change eye color or the number of fingers from scene to scene. At least, that’s the promise”.
He also claims that there will be support for more image types. “The model has been trained to work not only with photos but also with illustrations and cartoon characters,” he writes. “Video length of up to 1 minute. Native 1080p resolution. Multi-prompting for multi-shot (you can write several prompts to generate multiple scenes based on a single photo)”.
From the details leaking online, Veo 3.1 already sounds like it could be a great rival to OpenAI’s recently-released Sora 2. The ability to have multiple prompts and multi-shot video opens up the possibility of directing much more complex scenes, and the longer video length would also aid the process.
While we’re still waiting for Veo 3.1 to be added to Google Gemini, you can join the waitlist to get Veo 3.1 at Higgsfield and access it via the Envato and Imagine Art websites.
Follow TechRadar on Google News and add us as a preferred source to get our expert news, reviews, and opinion in your feeds. Make sure to click the Follow button!
And of course you can also follow TechRadar on TikTok for news, reviews, unboxings in video form, and get regular updates from us on WhatsApp too.