For several months now, Google Gemini has teased generative video capabilities, but the latest beta suggests those features are closer than ever. In Google app beta 16.11, Android Authority’s Abner Li found several strings that reveal a few details about the upcoming video generation features.
The first is a string that says “Get high-quality videos with Veo 2, Gemini’s latest video generation model.” Veo promises to create an eight-second video in about two minutes based on your idea. All users have to do is describe their vision in a few sentences. Don’t get too excited, though; it seems there will be usage limits, so you aren’t going to be creating feature-length films just yet.
Li found the following code:
While the limit isn’t clear, it’s obvious that Gemini will stop users after a certain point. You might be able to purchase credits for additional generative video, but again, this is just speculation. As for the “Toucan” reference, that’s the codename Google has used for testing Veo 2 inside Gemini.
Right now, Gemini users can take advantage of Imagen 3 for everything for everything from video, image, and audio generation, but its capabilities are limited. Veo 2 is capable of generating a larger range of visual styles while simulating real-world physics, according to Google’s DeepMind. It also is a much better choice “in terms of detail, realism, and artifact reduction.”
The presence of these code strings is a good hint that Veo 2 is inching ever closer to release, but until it’s official, anything could change. Google uses these beta versions of apps to test upcoming features and evaluate their validity. It’s possible the features will launch soon, but it’s equally as possible that a problem will arise that delays launch. For now, take this information with a grain of salt.
Read the full article here
