What is Gemini Omni Flash?
Google DeepMind's model card describes Gemini Omni as a step toward models that create and edit anything from any input, starting with video. Gemini Omni Flash takes text, images, audio and video as input and returns video with audio.
The Next Web reported its introduction at Google I/O in May 2026. Google opened a public preview in the Gemini API on June 30, 2026 with 10-second clips, followed by Gemini Omni 1.1 Flash on August 27, 2026 and 1.2 Flash on November 5, 2026.
What it can do
Google lists 1.1 Flash as adding first and last frame interpolation, scene extension, short video references, 360p drafts and upscaling to 1080p and 4K. 1.2 Flash makes clips of 3 to 15 seconds and improves lip-sync and dialogue between several speakers. The Gemini API documentation lists 1.2 Flash at 24 frames per second with native audio.
Google's deprecations page names Gemini Omni 1.2 Flash as the replacement for the Veo 3.1 preview models in the Gemini API, and Google Cloud says its outputs carry a SynthID watermark. The model card notes that consistency across edits, complex motion and accurate text remain hard.
On MakeMidas, Gemini Omni Flash makes 3- to 10-second clips at 360p, 720p, 1080p or 4K, in 16:9 or 9:16, from text, a first frame, or a first and last frame. Every clip comes with sound.













