I Tried Gemini Omni: Effortless Video Creation, Mind-Blowing Results

I am a frequent user of Google’s generative image AI, Gemini Nano Banana, which I use to merge photos, create illustrations, and participate in viral K-pop idol photo trends on social media.

This time, my curiosity turned to Google’s latest generative video AI, Gemini Omni, unveiled at Google I/O 2026 in Mountain View, California, last week.

Google describes Gemini Omni as an ‘any-to-any’ AI model, capable of accepting various inputs such as text, images, audio, and video, then transforming them into realistic and consistent new videos. The AI retains instructions, including characters, objects, and video atmosphere, though users may need to make ongoing revisions.

As someone who often dabbles in AI storyboards and illustrations, I was eager to test it myself.

However, Gemini Omni’s video creation feature is not available for free. Users must subscribe to Google’s paid AI services, with options starting at approximately Rp75,000 per month for Google AI Plus, Rp309,000 for Google AI Pro, and around Rp1.5 million for Google AI Ultra.

After subscribing, the “Create Video” option appears using Gemini 3.5 Flash. Google also provides a range of ready-to-use templates, including anime, comic book, talking pets, and cinematic vlog styles, making it easier for users to choose their desired video format.

However, I opted to test Gemini Omni’s raw capabilities using a concert storyboard previously created in ChatGPT. The storyboard consisted of nine frames depicting a woman attending a stadium concert. I simply uploaded the storyboard images to Gemini Omni and entered the prompt: “create a video based on this storyboard”.