Skip to content
Get started
GENERATION API CALLS
Video Generation

Google

This page is auto-generated from model configurations. Last updated: 2026-07-29.

This reference lists all available Google video generation models and their parameters. Use these parameter names when calling the Generation API.


Gemini Omni Flash generates 720p video with native audio from a text prompt or an input image.

Model ID: model_google-omni-flash

Capabilities: txt2video, img2video

LLM Markdown: https://app.scenario.com/api/models/model_google-omni-flash/markdown

ParameterTypeRequiredDefaultMinMaxAllowed ValuesDescription
promptstringNo----Describe the video you want to create — the scene, action, and mood. Optional if you provide a first-frame image instead.
imagefileNo----An optional image to animate. The video starts from this frame. Leave it empty to generate purely from your prompt.
referenceImagesfile_arrayNo----Up to 7 reference images of the subjects you want in the video.
durationnumberNo8310-The duration of the video in seconds.
aspectRatiostringNo16:9--16:9, 9:16The shape of the video — widescreen (16:9) or vertical (9:16).

Edit an existing video with natural-language instructions using Gemini Omni Flash — preserves motion while applying visual changes, with optional reference images (1–5) and native audio.

Model ID: model_google-omni-flash-edit

Capabilities: video2video

LLM Markdown: https://app.scenario.com/api/models/model_google-omni-flash-edit/markdown

ParameterTypeRequiredDefaultMinMaxAllowed ValuesDescription
promptstringYes----Describe the change you want to make to the video — for example, “make it look like winter” or “change the car to red.” The motion stays the same; only the look changes.
videofileYes----The video you want to edit.
referenceImagesfile_arrayNo----Optional reference images (1–5) injected into the edit for subject or look consistency.

Gemini Omni Flash generates subject-consistent 720p video with native audio from 1–7 reference images and an optional prompt.

Model ID: model_google-omni-flash-r2v

Capabilities: img2video

LLM Markdown: https://app.scenario.com/api/models/model_google-omni-flash-r2v/markdown

ParameterTypeRequiredDefaultMinMaxAllowed ValuesDescription
promptstringNo----Optionally describe the scene, action, or mood you want. Your reference subjects appear in the video whether or not you add a prompt.
referenceImagesfile_arrayYes----1 to 7 reference images of the subjects you want in the video.
durationnumberNo8310-The duration of the video in seconds.
aspectRatiostringNo16:9--16:9, 9:16The shape of the video - widescreen (16:9) or vertical (9:16).

Veo 3.1 is a realistic physics video model for simulating natural phenomena and physical interactions. Veo 3.1 can also generate sound and music.

Model ID: model_veo3-1

Capabilities: txt2video, img2video

LLM Markdown: https://app.scenario.com/api/models/model_veo3-1/markdown

ParameterTypeRequiredDefaultMinMaxAllowed ValuesDescription
promptstringNo----Describe your video
imagefileNo----Image used as the first frame of the video. Ideal images are 16:9 or 9:16 and 1280x720 or 720x1280, depending on the aspect ratio you choose. First Frame and Reference Images cannot be both set.
lastFrameImagefileNo----Last frame of the video to start generating from. When provided with an input image, creates a transition between the two images.
referenceImagesfile_arrayNo----1 to 3 reference images for subject-consistent generation (reference-to-video, or R2V). First Frame and Reference Images cannot be both set.
referenceImagesTypestringNoASSET--ASSET, STYLEThe type of the reference image, which defines how the reference image will be used to generate the video. ASSET is a reference image that provides assets to the generated video, such as the scene, an object, a character, etc. STYLE is A reference image that provides aesthetics including colors, lighting, texture, etc., to be used as the style of the generated video, such as ‘anime’, ‘photography’, ‘origami’, etc.
negativePromptstringNo----Description of what to discourage in the generated video
resolutionstringNo720p--720p, 1080pResolution of the generated video
generateAudiobooleanYestrue---Generate audio for the video
aspectRatiostringNo16:9--16:9, 9:16Aspect ratio for the generated video. If the aspect ratio of the input image does not match the selected video ratio, the model may crop the image to fit.
durationnumberNo8--4, 6, 8Video duration. With Reference Images, only 8s duration is supported.
seednumberNo----Use a seed for reproducible results. Leave blank to use a random seed.

Use Veo 3.1 to extend videos that you previously generated with Veo by 7 seconds and up to 20 times.

Model ID: model_veo3-1-extend-video

Capabilities: video2video

LLM Markdown: https://app.scenario.com/api/models/model_veo3-1-extend-video/markdown

ParameterTypeRequiredDefaultMinMaxAllowed ValuesDescription
promptstringNo----Describe your video
videofileYes----Input video to extend. Must be a clip in 16:9 aspect ratio, with a short-side resolution of 720p or 1080p.
generateAudiobooleanYestrue---Generate audio for the video
seednumberNo----Use a seed for reproducible results. Leave blank to use a random seed.

Veo 3.1 Fast is a faster, more affordable version of Veo 3.1, Google’s realistic physics video model for simulating natural phenomena and physical interactions. Veo 3.1 Fast can also generate sound and music.

Model ID: model_veo3-1-fast

Capabilities: txt2video, img2video

LLM Markdown: https://app.scenario.com/api/models/model_veo3-1-fast/markdown

ParameterTypeRequiredDefaultMinMaxAllowed ValuesDescription
promptstringNo----Describe your video
imagefileNo----Image used as the first frame of the video. Ideal images are 16:9 or 9:16 and 1280x720 or 720x1280, depending on the aspect ratio you choose. First Frame and Reference Images cannot be both set.
lastFrameImagefileNo----Last frame of the video to start generating from. When provided with an input image, creates a transition between the two images.
referenceImagesfile_arrayNo----1 to 3 reference images for subject-consistent generation (reference-to-video, or R2V). First Frame and Reference Images cannot be both set.
negativePromptstringNo----Description of what to discourage in the generated video
resolutionstringNo720p--720p, 1080pResolution of the generated video
generateAudiobooleanYestrue---Generate audio for the video
aspectRatiostringNo16:9--16:9, 9:16Aspect ratio for the generated video. If the aspect ratio of the input image does not match the selected video ratio, the model may crop the image to fit.
durationnumberNo8--4, 6, 8Video duration. With Reference Images, only 8s duration is supported.
seednumberNo----Use a seed for reproducible results. Leave blank to use a random seed.

Veo 3.1 Lite is a lite version of Veo 3.1, Google’s realistic physics video model for simulating natural phenomena and physical interactions. Veo 3.1 Lite can also generate sound and music.

Model ID: model_veo3-1-lite

Capabilities: txt2video, img2video

LLM Markdown: https://app.scenario.com/api/models/model_veo3-1-lite/markdown

ParameterTypeRequiredDefaultMinMaxAllowed ValuesDescription
promptstringNo----Describe your video
imagefileNo----Image used as the first frame of the video. Ideal images are 16:9 or 9:16 and 1280x720 or 720x1280, depending on the aspect ratio you choose.
lastFrameImagefileNo----Last frame of the video to start generating from. When provided with an input image, creates a transition between the two images.
negativePromptstringNo----Description of what to discourage in the generated video
resolutionstringNo720p--720p, 1080pResolution of the generated video
generateAudiobooleanYestrue---Generate audio for the video
aspectRatiostringNo16:9--16:9, 9:16Aspect ratio for the generated video. If the aspect ratio of the input image does not match the selected video ratio, the model may crop the image to fit.
durationnumberNo8--4, 6, 8Video duration
seednumberNo----Use a seed for reproducible results. Leave blank to use a random seed.