Google launched Veo 3.1, a significant update to its video generation tool for artificial intelligence. The new feature doesn’t come alone: ​​along with it, Flow — the company’s video editor — gains granular editing features that allow you to add or remove objects from scenes, adjust lighting and create cinematic transitions between frames.​

Since May, when Flow was launched, the platform has generated more than 275 million videos. Now, Veo 3.1 elevates audiovisual quality and offers more refined narrative control for creators.

Or what mudou I don’t see 3.1

The main evolution is richer and more contextual audio. Veo 3.1 improves the quality of sound generation that already existed in Veo 3, offering more detailed audio with better lip sync. Additionally, Google has added audio generation to three Flow features that previously produced silent video: Ingredients to Video, Frames to Video, and Scene Extension. The model also adheres better to prompts — that is, it understands and more accurately executes what the user describes in text.

Another advance is in converting images into video. Veo 3.1 can maintain character consistency between multiple scenes when fed with reference images, a valuable resource for those who need to create serialized content or branded content.

The tool also allows you to use up to three reference images to guide video generation — useful for maintaining a character’s visual identity or applying a specific style. Realism has improved: textures have become closer to reality and the understanding of cinematographic styles has become more sophisticated.​

Editing tools that make a difference

Within Flow, three features gain audio for the first time. THE Ingredients to Video allows you to combine multiple reference images — characters, objects and scenes — to create a complete scene with audio.​

Already the Frames to Video generates a video connecting an initial and final image, filling the movement between them with fluid transitions. Finally, the Extend zooms in on existing clips based on the last second of the original video, allowing you to create long takes of up to a minute.

Flow even allows you to add objects to existing videos, integrating them into the visual style of the scene. Soon, it will be possible to remove unwanted objects, with automatic adjustment of the background so that it looks like the element was never there.​

Where to access and how much does it cost

Veo 3.1 is available in the Gemini app (with limited free access), in Flow, and via the Gemini and Vertex AI APIs. For developers, the model can be accessed in both Google AI Studio and Vertex AI, with a faster version called Veo 3.1 Fast.​

Free Gemini users have a limited number of generations per day, while subscribers receive priority and additional credits. For Flow, a Google One AI Pro or AI Ultra subscription is required. Developers who use the APIs operate under a “paid preview” regime, with pricing similar to Veo 3

In addition to official Google platforms, Veo 3.1 can also be accessed through third-party tools such as Higgsfield AI, which offers an alternative interface for generating videos with the model.

Practical examples show the potential

On social networks, users are already sharing several videos generated with the tool. We separated some. Check it out:

~

Source: https://www.hardware.com.br/tecnologia/veo-3-1-videos-impressionantes-ia-google/



Leave a Reply

Related News

Intel gives up selling its NEX division, focused on networks and edge computing

December 5, 2025

NVIDIA restores PhysX support to RTX 50 for popular games

December 5, 2025