Google Gemini Docs Reveal 'Gemini Omni' Video Generation Model
GeminiApp · x · 2026-08-07
Google has updated its Gemini Apps Help documentation to introduce Gemini Omni, its latest AI video generation model. The model allows users to create and edit videos in minutes using a conversational interface, blending text, images, and video inputs.
Key features include:
- Video to Video Edit: Upload a video and instruct the model to apply edits.
- Multi-turn editing: Continually edit a video within a single conversation, with support for image or video references in each turn.
Currently, the feature requires a personal Google AI plan subscription and is unavailable to users under 18.
More from Multimodal
- MiniMax Video Generation: How to Reliably Swap Character Identities? — Ill_Key_7122 · 2026-08-07
- MiniMax H3 Image-to-Video: Can We Input Character and Environment Separately? — Emergency-Board-3042 · 2026-08-07
- ComfyUI Node: Prompt Builder for MiniMax H3 Reference 2 Video — eggplantpot · 2026-08-07
- MiniMax Video Generation Tested: $0.13/sec for 2K Resolution — MaziyarPanahi · 2026-08-07
- Spooky MiniMax Video Demo: AI-Generated Creepy Vibes — alanskimp · 2026-08-07
- Viral AI Series Zephyr Open-Sources All Prompts and Assets After 1.3M Views — dr_cintas · 2026-08-07