In a bid to expand its capabilities, Google is introducing video generation through its Veo 2 model in the Gemini app for Advanced subscribers.
Revealed late last year, Veo 2 promises “smooth character movements, lifelike environments, and intricate visual intricacies across various subjects and styles,” alongside a level of “cinematic realism” derived from a nuanced understanding of real-world physics and human dynamics.
Within the Gemini app, Veo 2 can produce eight-second video clips at a resolution of 720p. The output will be an MP4 file in a 16:9 landscape format, with an option to share via a g.co/gemini/share/ link. To provide your prompt, simply select Veo 2 from the model dropdown on both web and mobile platforms.
Just outline your desired scene: “Providing more details in your description gives you greater control over the end video.” The generation process takes approximately 1-2 minutes.
This feature opens a plethora of creative avenues, allowing your imagination to flourish with unique combinations, exploring diverse visual styles from realism to fantasy, or swiftly visualizing short narrative ideas.
Here are some sample prompts and their respective outputs (which we generated):
- “A small puppy frolicking through a garden blanketed in thick snow under the gentle morning sunlight.”
- “A kitten discovers the beach, particularly near the water, only to be startled by a crashing wave nearby.”
- “An aerial view of a grassy cliff overlooking a sandy beach where waves crash on the shore, featuring a prominent sea stack rising from the ocean near the beach, illuminated by the warm, golden light of sunrise or sunset, showcasing the dramatic elevation change and the tranquil beauty of the Pacific coastline.”
Regarding safety, each frame includes a SynthID digital watermark.
Exclusively available for Gemini Advanced subscribers ($19.99 per month), there is a “monthly limit” on the number of videos you can create, with Google notifying users when they approach that limit. This feature is being rolled out globally across all languages supported by Gemini, starting today and will be fully accessible in the upcoming weeks.
Additionally, Google One AI Premium subscribers are now able to access Veo 2 generation through Whisk. Launched in December, this Google Labs experiment allows users to “prompt with images” rather than text. The new feature, “Whisk Animate,” enables you to convert the images you create into vibrant eight-second videos using Veo 2.
