Gemini Omni Flash
Gemini Omni Flash takes text, images, audio and video in one prompt and outputs editable video, folding the Veo line into Gemini itself.
Announced at I/O 2026 as 'create anything from any input, starting with video'; conversational multi-turn edits, character consistency, synced audio, SynthID. Rolled out to Plus/Pro/Ultra and free in YouTube Shorts; clips capped at 10 s. API preview 2026-06-30, GA 2026-08-27 (360p-4K).
- Date
- Tuesday, 19 May 2026
- Lab
- Google DeepMind
- Kind
- model
- Access
- closed API
Google's I/O page confirms Omni Flash, its multimodal scope and rollout; the 10-second cap and 'Veo folded into Gemini' framing come from a third-party site (vo3ai.com). Dates for API preview/GA from the Gemini API changelog.
Sources
- blog.google/innovation-and-ai/technology/ai/google-io-2026-all-our-announcements/
- ai.google.dev/gemini-api/docs/changelog
- www.vo3ai.com/gemini-omni
This record was checked against its sources on 6 October 2026. How we check