Model guide

Gemini Omni Flash: Google’s Conversational AI Video Model

Gemini Omni Flash is Google’s model family for fast conversational video generation and editing, with video as its documented API output.

Explore Story321 video tools Open in Google AI Studio

Text to video

Image to video

Video editing

What is Gemini Omni Flash?

Gemini Omni Flash is a Google AI model family built for fast conversational video generation and editing. Google’s Gemini API documentation identifies Gemini Omni 1.1 Flash as the current stable API model and lists video as the API output modality. The Gemini Omni Flash API can work from text, images, or supported video inputs. Google documents workflows for creating a video from a prompt, animating an image, editing or extending an uploaded clip, and refining a result through a stateful interaction. This is an independent model guide. Story321 does not claim that Gemini Omni Flash powers its video tools.

Documented capabilities

Text-to-video

Create a video from a text prompt through the Gemini API workflow documented by Google.

Image-to-video

Use an image as the starting point for an animated video result.

Conversational editing

Use the Interactions API and a previous interaction ID to continue refining an earlier result.

Video editing and extension

Google documents uploaded-video editing and appending an extension to a qualifying clip.

Frame interpolation

Generate a transition video from supplied first and last frames.

Multiple output options

Official documentation lists 360p, 720p, 1080p, and 4K output options, subject to the selected workflow and availability.

Current API facts

Features	Gemini Omni Flash
Stable API model	gemini-omni-1.1-flash
Preview API model	gemini-omni-flash-preview
Documented input	Text, image, and video; uploaded video for editing or extension is limited to qualifying clips of up to 10 seconds.
Documented output	Video
Documented video duration	3 to 10 seconds
Documented frame rate	24 FPS
Documented context window	1,048,576 tokens

Where it can fit

Prompt-led short clips

Prototype brief video concepts from a written scene, action, or visual direction.

Still-image animation

Turn a supplied image into a short moving shot when an image-to-video workflow is appropriate.

Iterative video direction

Continue a conversation about a generated result instead of starting every edit from scratch.

Short-clip revisions

Edit a qualifying uploaded clip or append a continuation when the official workflow supports it.

Access and availability

Google documents Gemini Omni Flash in the Gemini API and provides an AI Studio entry point for Gemini Omni 1.1 Flash. Check Google’s current documentation before building a workflow because model identifiers, regional availability, and supported features can change. The model name may refer to the wider Gemini Omni family in Google materials, while Gemini Omni 1.1 Flash is the stable API model named in the current model documentation.

Important limitations

Documented limitation

Google lists restrictions for some editing and extension workflows by region, including the EEA, Switzerland, and the UK.

Documented limitation

Uploaded-video editing and extension are limited to supported clips of up to 10 seconds; extension is appended to the end of a video.

Documented limitation

Google’s current API guide lists voice editing, audio references, and multi-video workflows as unsupported or unavailable in the described workflow.

Documented limitation

The Gemini Omni Flash model card notes limitations around edit consistency, complex motion, and accurate text rendering.

Official sources

Gemini Omni Flash model documentation

Google’s model page lists the stable and preview API model identifiers, documented modalities, duration, resolutions, frame rate, and context window.

Google AI for Developers

Gemini Omni API guide

Google’s workflow guide covers text-to-video, image-to-video, stateful editing, frame interpolation, video editing, extension, and current limitations.

Google AI for Developers

Gemini Omni Flash model card

Google DeepMind’s model card describes intended use and published limitations, including consistency, motion, and text-rendering challenges.

Google DeepMind

Gemini Omni overview

Google DeepMind’s overview describes Gemini Omni as a model for creating and editing videos with conversational direction.

Google DeepMind

Gemini Omni Flash FAQ

Is Gemini Omni Flash a text-to-video model?

Google documents a text-to-video workflow for Gemini Omni Flash. Its current model page lists text, image, and video input with video output in the API.

What is the stable Gemini Omni Flash API model?

Google’s current model documentation lists gemini-omni-1.1-flash as the stable API model and gemini-omni-flash-preview as a preview model.

Can Gemini Omni Flash edit video?

Google documents editing and extending supported uploaded clips, plus stateful conversational refinement. Availability and workflow restrictions apply.