Omni Flash is an independent AI video generation studio that runs entirely in the browser. You describe a scene, drop in a reference still or paste a rough script, pick a model, and the engine returns a short clip — several plans advertise up to roughly 30 seconds per generation with support for many reference images and up to 4K masters. The selling point is that audio is produced together with the picture instead of being bolted on afterwards, so lip-sync and ambience land on the same beat as the visuals.
Beyond raw generation, the product leans on director-style controls. You steer the camera with everyday film vocabulary (tracking shots, racks, close-ups), save a colour grade and motion profile as a reusable preset, and refine a finished take by typing corrections such as holding a close-up for two extra beats. Character locking keeps a face, outfit or branded object stable across separate generations, which is what makes the output usable for episodic or campaign work rather than one-off clips.
Worth calling out: Omni Flash is a third-party platform with no official affiliation to Google or its similarly named generative video research. Its pricing page also indicates that in addition to its own Omni models, the workspace offers credits for external video models, positioning it as an aggregator as much as a single model. Every render is described as coming with a commercial-use licence, watermark-free masters and provenance metadata.
