Gemini Omni 1.1 Flash Image-to-Video API by Google

Gemini Omni 1.1 Flash — Image to Video

Model ID: google/gemini-omni-1.1-flash/image-to-video

Gemini Omni 1.1 Flash is Google DeepMind's natively multimodal model for video generation and editing. This variant animates a still image according to a text prompt, and — new in 1.1 — accepts an optional last frame so the model interpolates the whole shot between two keyframes you choose.

Overview

Gemini Omni Flash is a single transformer that processes text, images, audio, and video simultaneously rather than stitching separate modality-specific pipelines together. The result is output that is more cohesive, more consistent across turns, and more controllable than a chained generate-then-dub workflow.

What sets it apart from earlier video models is that it natively generates audio with every video — dialogue, ambience, music, and sound design are produced together with the picture rather than added afterward. The model is grounded in Gemini's real-world knowledge, so it reasons about physics, narrative logic, culture, and visual composition to produce results that feel intentional and cinematic. Every generated clip carries an imperceptible SynthID watermark alongside C2PA Content Credentials for provenance verification.

What's new in 1.1

Version 1.1 is a control-and-fidelity upgrade over the first Omni Flash release:

• Flexible resolution output — new 4K and 1080p rendering for finished work, plus a fast 360p draft mode for previewing a shot before committing to a full-quality render.
• Longer continuous generation — shots can be extended segment by segment to a total of up to 40 seconds of coherent video.
• Precise shot start and end control — first and last frame can both be supplied, producing smoother camera moves, scene transitions, and seamless loops.
• Character and style consistency — a new video reference capability (clips of up to 3 seconds) substantially improves subject and art-direction stability.

How the five endpoints relate

AtlasCloud exposes Gemini Omni 1.1 Flash through five model IDs — text-to-video, image-to-video, reference-to-video, video-edit, and video-extend. All five route to the same underlying gemini-omni-1.1-flash model and differ only by the input modality they accept, corresponding to the model's task parameter (text_to_video, image_to_video, reference_to_video, edit, extend). This endpoint maps to image_to_video.

Inputs

This variant takes a text prompt and a starting image, with an optional ending image.

• Prompt — natural-language description of the motion, camera language, style, and audio to apply, up to 20,000 characters.
• image (required) — the first frame of the generated video. PNG, JPEG, JPG, or WebP, up to 20 MB, supplied as a public URL or a base64-encoded image.
• last_image (optional) — the last frame of the generated video. When present, the model interpolates the shot between image and last_image. Setting it requires image to also be set.

Keyframe control is the headline reason to use this endpoint over text-to-video: pinning both ends of a shot gives you deterministic composition at the cut points, which is what makes complex camera moves, scene-to-scene transitions, and seamless loops reliable.

Key Capabilities

• Image animation — bring a photograph, render, or illustration into motion while preserving its subject and composition.
• First / last frame interpolation — supply both ends of the shot and let the model generate a smooth path between them.
• Native audio generation — every clip is rendered with a synchronized soundtrack (speech, music, effects) driven by your description.
• Draft-to-final resolution ladder — iterate cheaply at 360p, then re-render the shot you like at 720p, 1080p, or 4K.
• Cinematic prompt understanding — shot framing, camera movement, lens character, lighting, and grade all respond to plain-language direction.
• World-grounded realism — physics, motion, and scene dynamics informed by Gemini's real-world knowledge.
• Adjustable reasoning — the thinking_level control trades latency for quality on complex prompts.
• Reproducible results — set a fixed seed to reproduce or iterate on a specific generation.

Input Parameters

A single generation call produces at most 10 seconds. To reach the model's 40-second ceiling, chain the result into the video-extend endpoint.

Prompting Tips

• Describe what changes, not what is already visible. The input image establishes the subject and composition; the prompt should carry motion, camera, and audio.
• For a seamless loop, pass the same asset as image and last_image and describe the round-trip motion in between.
• For a scene transition, make image and last_image the two compositions you want to cut between and let the model invent the connective camera move.
• Draft at 360p while you tune wording, then re-run with a fixed seed at 1080p or 4k.

Use Cases

• Photo and artwork animation — put stills, product renders, or illustrations into motion.
• Keyframe-driven shots — lock the opening and closing composition of a shot and generate the move between them.
• Seamless loops — background plates, banners, and ambient visuals that repeat without a visible cut.
• Scene transitions — bridge two distinct compositions with a single generated camera move.
• Product motion — animate a hero shot for advertising or storefront use.

Pricing

Billing is based on the duration of the generated video, charged at a per-second rate that depends on the requested resolution.

SKUs

Formula

cost = max(3, duration) × resolution_rate

Where resolution_rate is $0.043 for 360p, $0.11 for 720p, $0.16 for 1080p, and $0.31 for 4k.

Notes and examples

• Billing is per second of generated video, with a 3-second minimum — a shorter request is still billed as 3 seconds.
• duration defaults to 10 and resolution defaults to 720p when not supplied.
• Supplying image or last_image adds no separate charge; input frames are not billed.
• A 10-second 720p video costs 10 × $0.11 = $1.10.
• The same 10 seconds as a 360p draft costs 10 × $0.043 = $0.43.
• A 10-second 1080p video costs 10 × $0.16 = $1.60.
• A 10-second 4K video costs 10 × $0.31 = $3.10.
• A 3-second 720p video costs 3 × $0.11 = $0.33.