Newsletter

Google stated that Google Vids is now offering free 1080p video generation and zooming capabilities to any users who possess an account with Google or Workspace. The access address is vids.new.

This feature operates on Gemini Omni 1.1 Flash, and it includes enhancements such as extended scenario support, precise segment duration control, and high-definition output. Each video will have an invisible SynthID watermark.

This free opening coincides with the complete closure of Sora API on the same day as OpenAI. Previously, Google had already reduced the price of its own Veo 3.1 Lite API to 0.05 US dollars per second.

Google stated that any user with an account with the identifier Google can now generate full HD videos in Full HD quality for free using the company's browser-based video editor Google Vids. This feature was announced on September 23rd.

This tool runs on Gemini Omni 1.1 Flash. It is a AI model that can convert text prompts, photos, audio, or existing clips into new video materials. Previously, such generation capabilities were mostly available only with a subscription.

No longer necessary. Users simply need to go to vids.new, log in with any Google account, and they can start generating videos at a resolution of 1080p. Google claims that this is consistent with the high-definition resolution used by most streaming platforms, and there is no need to pay for it.

We conducted tests, and the generation process is quite straightforward. However, we are only able to generate videos at a resolution of 720p, whether in landscape or portrait format.

Google has also added several editing controls that previously required paid tools to use. The extended scene function allows for stretching of segments while maintaining consistency in lighting, characters, and scenery, rather than having the image change suddenly in the middle.

Users can also set precise duration for scenes to align the clips with the script or voiceover, and they can resize earlier AI clips with lower resolution to 1080p without having to regenerate them from scratch.

Each video generated by this tool contains SynthID embedded within it. This is a watermark from Google's DeepMind, which is invisible to the human eye but can be detected by machines and is directly written into the video frames. Anyone can use Google's SynthID Detector to check whether a video was generated by AI. This is an independent tool designed to identify content that has been impersonated as genuine AI-generated material.

Its speed is also very fast. We generated a short video in less than 30 seconds, which is slightly faster than the usual time it takes when using Google Flow.

In the past, creating a decent high-definition video meant paying for software, hiring a editor, subscribing to material videos, or owning a high-performance computer capable of running open-source video models. Now, all that's required is a free browser tab, and this has changed who can afford to produce video content, with both advantages and disadvantages.

Not every account is entitled to the same amount of free generation quotas. Individual Google accounts will receive standard quotas, while Google's paid AI plan allows for unlocking more quotas. Subscribers of Workspace Business and Enterprise will have access to larger generation pools and can manage these quotas within their teams through administrative controls.

The timing is hard not to notice. OpenAI shut down its consumer-facing Sora app and website in April 2026, and its developer API – the last part that was still in operation – also completely ceased services today, coinciding precisely with the same day that Google made Vids available for free.

Google is also accelerating its efforts in terms of pricing. Recently, the company reduced the API fee for its Veo 3.1 Lite model to 0.05 US dollars per second, putting pressure on its competitors in the developer community; at the same time, Chinese rivals such as Kling and Hunyuan Video are also continuously exerting pressure on prices from the other side.

Google stated that the next supporting feature, Gemini 3.8 Flash-Lite text-to-speech, will soon be launched, allowing written scripts to be converted into dubbed narrations in over 100 languages.