Sam Witteveen · 2026-06-30 · notable
Sam Witteveen: 'Introducing the Gemini Omni Flash API'
Sam Witteveen walks through the Gemini Omni Flash API — Google DeepMind's multimodal video-generation model, now reachable from code as Google opens its developer rollout.
A hands-on look at Google DeepMind's Gemini Omni Flash, now exposed as an API for developers and enterprises.
What is it?
Sam Witteveen's new video covers the Gemini Omni Flash API. Gemini Omni Flash is Google DeepMind's multimodal model that generates and edits high-quality video from any mix of text, image, audio, or video inputs. The model was first published 19 May 2026 and is now rolling out to developers and enterprise customers via the Gemini API.
How does it work?
Gemini Omni Flash is a transformer with native multimodal support for text, vision, video, and audio. Witteveen, an AI educator best known on YouTube for hands-on API walkthroughs, picks the developer angle for this video: how to call the model from code, what inputs and outputs look like, and where Gemini Omni Flash sits next to the rest of the Gemini family.
Why does it matter?
Most coverage of Gemini Omni Flash so far has been demo reels in the Gemini app and YouTube Shorts. A creator-led API walkthrough is the first practical look at how developers will actually wire the model into their own apps, agents, or video pipelines now that the API rollout has started.