SuperAPI AI Directory

Gemini Omni

Gemini Omni is an AI video model reportedly launched by Google. According to its official description, it can generate video clips up to 1080p and up to 30 seconds long, and supports editing videos conversationally in Gemini chat and remixing existing footage without timeline operations. The site also offers a library of over 100 structured prompt templates covering categories such as cinematic, advertising, and natural scenery, and supports uploading images for motion generation (PNG, JPG, WebP, max 10MB, estimated to consume 5 credits). Its release information states that the model was launched at Google I/O 2026. Apart from the above, information on pricing, platform availability, and specific usage steps is not yet provided.

Open website

About this tool

Purpose

Gemini Omni is introduced as Google's AI video model. Official text states that it was released at Google I/O 2026, with the goal of changing how AI videos are generated and edited. In addition to introducing the model's capabilities, the site positions itself as a prompt template library and example showcase platform to help users quickly get started with video generation.

Main Capabilities

According to the official description, Gemini Omni has the following capabilities:

  • Generate 1080p video clips, with each segment up to 30 seconds long;
  • Edit videos conversationally in Gemini chat (in-chat editing);
  • Remix existing video footage (video remixing), which the official description says requires no timeline;
  • Support uploading images for animation generation: PNG, JPG, and WebP can be uploaded, with a maximum file size of 10MB, an estimated consumption of 5 credits, and a required motion description; the generated result will be displayed in the page's generation area.

How to Use

The site provides over 100 structured prompt templates, browsable by category, which can be copied directly and pasted into Gemini Omni to generate videos. The official description emphasizes that Gemini Omni can follow instructions precisely, so the templates lock in five key elements: camera move, subject, action, environment, and style. Users can also choose to generate directly in the app and test templates using the provided AI video generation credits.

The examples on the site are displayed categorized by theme, including:

  • Cinematic: such as "Lone Hiker Revealed at the Edge of a Colossal Canyon," "180-Degree Arc Shot Around a Singer on Stage," "Young Woman Gazing Out a Bus Window at Night," "Melancholic Subject on Spinning Playground Wheel," etc.;
  • Ads & Commercial: such as "Insect Simulator Macro Survival Thriller," "Surreal Tennis Ball Loop." The example prompts are in structured JSON format, including fields such as aspect_ratio, resolution, framerate, style, camera_motion, and lighting;
  • Nature & Landscape: such as "Arctic Fox Chase Drone Shot," "ASMR Glass Strawberry Macro Cutting."

Each example is equipped with "Copy Prompt" and "Try It Yourself" entries.

Suitable Scenarios

Suitable for creators who need to quickly produce short video materials, including cinematic narrative short films, advertising and commercial content, nature and landscape shots, macro close-ups, ASMR-style videos, and more. The official description says it is suitable for creators who want to consistently produce videos every week.

Limitations and Information Not Yet Provided

  • Information on pricing, subscription plans, and the specific price and acquisition method of credits is not yet provided;
  • Information on available platforms, regional restrictions, API access, commercial licensing, etc. is not yet provided;
  • Apart from 1080p, 30 seconds, 10MB image uploads, and the 5-credit estimate, other technical specifications (such as frame rate limits, export formats, watermark policies, etc.) are not yet provided;
  • Statements such as "released at Google I/O 2026" come from the site's official text and are recorded only as factual information.