Google expands creative controls in its Gemini-powered Google Flow video editor

Google is rolling out new “creative control” features in Google Flow, a video editing experience powered by Gemini, as the company pushes deeper into AI-assisted content creation tools for consumers and professionals. In a post tied to its August 27, 2026 rollout, Google said the updates are now available in Google Flow for users looking to move from rough cuts to production-ready video with fewer manual steps.

The headline addition is what Google calls start and end frames, designed to help maintain consistency in both characters and narrative when creators transition between frames. Google’s framing is that edits often drift away from the intended look or story as people experiment with cuts, and the new controls aim to keep those elements steadier throughout the sequence.

On the output side, Google Flow now supports exports at 1080p and 4K resolution, positioning the tool for “high-end digital, social, or broadcast editing workflows,” according to the company. That matters because AI-assisted editors often focus on quick drafts, and moving to higher-resolution exports indicates an effort to make the platform more viable for more demanding use cases.

Google also described a draft-and-upscale workflow intended to speed up experimentation. Users can test concepts and compositions at a lower, “faster” 360p resolution before committing to a full-resolution render. Once a clip meets the creator’s expectations, Google said users can download it at 720p.

The company highlighted the mobile angle, noting that the approach is “particularly helpful” in the Google Flow app. Google said users can draft videos on their phones and then upscale versions, implying that the product is designed to reduce the time and compute cost of iterating on ideas during on-the-go editing.

The rollout builds on earlier work Google referenced from Google I/O, where it said it launched Gemini Omni Flash in Google Flow. In its update, Google said it is now applying changes via Gemini Omni 1.1 Flash, aiming to provide “more creative control” and help users craft “more polished and production-ready videos.”

From a business perspective, the updates underscore how Alphabet’s technology units are extending the competitive field of AI video tooling. Google Flow sits alongside a broader push across AI models and creative software, where the differentiator is increasingly the ability to translate text, prompts, or selected frames into editing outcomes while keeping results coherent.

For Alphabet, this also reflects a strategy to connect frontier models to consumer-facing products where usage can scale quickly. Even without disclosed user metrics in the announcement, the pattern is consistent with how Google tends to productize model capabilities: start with a limited set of creative tasks, then broaden controls and output quality to make the workflow more end-to-end.

Google did not provide details in the announcement about how broadly the start-and-end-frame control behaves across different content types, whether it supports specific aspect ratios or editing primitives beyond transitions, or what limitations may apply to complex scenes. It also did not specify whether the 1080p or 4K export options require particular plan tiers, regional availability, or hardware constraints.

Looking ahead, creators and industry watchers are likely to focus on whether Google Flow can deliver reliable consistency controls as edits become more complex, and how quickly Google iterates on export quality and render performance. Another near-term question is how Gemini Omni 1.1 Flash improvements translate into user time saved, particularly for mobile-first editing workflows that rely on draft-and-upscale steps. As Google continues rolling features into Google Flow, the competition will likely track which platform can best balance speed, coherence, and export-ready output.