Gemini Advanced can now create videos, and they are incredible.

      Google has introduced a new and exciting feature to its AI personal assistant and chatbot, Gemini Advanced. With just a text prompt, Gemini can now generate an 8-second animated video, bringing your words to life in an astonishing way. This feature utilizes Veo 2, a video model unveiled in late 2024, designed to produce realistic videos that show a deep comprehension of human movements, real-world settings, and various lens types.

      According to Google, creating videos with Gemini and Veo 2 is straightforward. "Simply describe the scene you wish to create—whether it's a short story, a visual idea, or a specific moment—and Gemini will bring your concepts to reality. The more detailed your description, the more control you have over the resulting video. This opens up numerous fun creative possibilities, allowing your imagination to run wild to envision imaginative combinations, experiment with different visual styles from realism to fantasy, or quickly communicate brief visual notions."

      The videos generated by Google Gemini are eight seconds long, produced in 720p resolution, and have a 16:9 aspect ratio. You can download the MP4 files from your chat with Gemini or share them directly to Facebook, Reddit, LinkedIn, or X from the feed. Videos can also be shared via a public link, and you can view some of the videos created with Veo 2 through the links in the following section.

      **Trying Out Google Gemini’s Video Creation**

      If you've previously attempted to create videos with AI, you already understand that the level of detail in your description often determines how closely the result aligns with your vision. Google Gemini functions similarly, but it excels at filling in gaps when your prompt is vague, and it also offers some unique “ideas” of its own.

      For instance, I provided a simple, intentionally brief prompt for a video depicting "A K-pop girl group performing on stage in a large stadium. Thousands of fans are watching and singing along, with most holding light sticks in support." The resulting video is remarkable, featuring the main singer with a distinct K-pop idol look, and the stadium is suitably crowded. I particularly appreciated the Korean text at the bottom of the screen, enhancing the scene and giving the impression that it was a clip from a live-streamed performance.

      In another instance, I used a much more detailed description to create a countryside scene where a man and a cat stroll along a lane bordered by abandoned farm structures, with UFOs hovering over distant fields. The final video turned out excellent, capturing the leafy lane I envisioned, the cat's movements, and the appearance of the UFOs as the camera moved past the trees. These videos were made rapidly from simple prompts, and it was delightful to see the capabilities of Veo 2. Spending more time on crafting your scene reveals even more potential.

      **How to Access Video Creation in Gemini**

      The Veo 2 video creation feature in Google Gemini is available to subscribers of Gemini Advanced, which costs $20 per month or 19 British pounds per month. It can be found in a new drop-down menu, allowing you to switch between different models—2.0 Flash, 2.0 Flash Thinking, 2.5 Pro, and Deep Research with 2.5 Pro. Veo 2 can be used on both desktop and mobile, and Google has indicated that the feature will begin rolling out today and will continue to be available in the upcoming weeks.

      If you have a Google One AI Premium subscription ($20 per month) and access to Google's Whisk tool, you can use images to assist in video creation. Currently, images cannot be added to Veo 2 for style or appearance guidance. Whisk, which was previously available only in select regions, is now accessible worldwide to subscribers.