# Gemini Omni Video
## OpenAPI Specification
```yaml
openapi: 3.0.1
info:
title: ''
description: ''
version: 1.0.0
paths:
/api/v1/jobs/createTask:
post:
summary: Gemini Omni Video
deprecated: false
description: >+
## Create Task
Use this endpoint to create a new multimodal video generation task.
After submission, use the unified query endpoint to check task progress and retrieve results
::: tip[]
For production use, we recommend providing the `callBackUrl` parameter
so your service can receive completion notifications instead of polling
for task status.
:::
## Related Resources
Explore all available models and capabilities
Check account credits and usage
**Total Quota Overview:**
The system utilizes a quota-based system for content uploads. The total
available capacity is **7 units**. Different parameters consume
different amounts of this quota. Please ensure the sum of all consumed
units does not exceed 7.
**Quota Consumption Rules:**
1. **Images (`image_urls`)**: Each image consumes **1 unit**.
2. **Videos (`video_list`)**: Each video consumes **2 units**.
* *Note: Maximum of 1 video per request.*
3. **Character IDs (`character_ids`)**: Each ID consumes **1 unit**.
* *Note: Maximum of 3 character IDs per request.*
**Constraint Formula:**
`(Number of Images) + (Number of Videos × 2) + (Number of Character IDs)
≤ 7`
**Example Scenarios:**
* If 1 video (2 units) and 3 Character IDs (3 units) are uploaded, you
can upload a maximum of 2 additional images.
* If no video or Character IDs are provided, you can upload up to 7
images.
operationId: gemini-omni-video
tags:
- docs/en/Market/Video Models/Gemini Omni
parameters: []
requestBody:
content:
application/json:
schema:
type: object
required:
- model
- input
properties:
model:
type: string
enum:
- gemini-omni-video
default: gemini-omni-video
description: >-
The model name used for generation. This field is required.
This endpoint must use the `gemini-omni-video` model.
examples:
- gemini-omni-video
callBackUrl:
type: string
format: uri
description: >-
Callback URL for task completion notifications. Optional
parameter. If provided, the system will send a POST request
to this URL when the task completes, whether it succeeds or
fails. If omitted, no callback notification will be sent.
examples:
- https://your-domain.com/api/callback
input:
type: object
description: Input parameters for the multimodal video generation task.
properties:
prompt:
type: string
description: >-
Video prompt used to describe the target content, style,
camera language, or character actions in the generated
video.
examples:
- >-
Create a futuristic night city short film with a slow
push-in shot as the character walks out from a
neon-lit street.
maxLength: 20000
image_urls:
type: array
items:
type: string
format: uri
description: >-
Array of image URLs. You can provide one or more
reference images for characters, scenes, styles, or
storyboard guidance.
Image limits:
- Each file must be no larger than `20MB`
- Use publicly accessible image URLs
- Max 7 images
examples:
- - https://example.com/assets/scene-1.png
- https://example.com/assets/scene-2.png
audio_ids:
type: array
items:
type: string
description: >-
Array of audio IDs generated by the `gemini-omni-audio`
endpoint. Useful for narration, dialogue, music, or
audio guidance in the generated video. Max 3 items.
examples:
- - audio_01hx8p0demo
video_list:
type: array
items:
type: object
required:
- url
- start
- ends
properties:
url:
type: string
format: uri
description: >-
Video URL. Each source video file must be no
larger than `100MB` and no longer than `30s`.
start:
type: number
minimum: 0
description: Start time in seconds.
ends:
type: number
minimum: 0
description: >-
End time in seconds. It should be greater than
`start`.The difference between the end time and
the start time must not exceed 10s
x-apidog-orders:
- url
- start
- ends
description: >-
Array of video clips. Each item defines a source video
and the trim range to use during generation.
Video limits:
- Each file must be no larger than `100MB`
- Video duration must not exceed `30s`
- `ends` should be greater than `start`
- The difference between the end time and the start time
must not exceed `10s`.
- Max 1 items. Equal 2 images
examples:
- - url: https://example.com/assets/source-video.mp4
start: 0
ends: 10
character_ids:
type: array
items:
type: string
description: >-
An array of character IDs generated by the
`gemini-omni-character` API. Used to provide character
appearance, identity, or person references for the
video. Each character_id uses 1 image slot. The base
limit is 7 image slots; if video_list is also provided,
video_list uses 2 image slots, so character_ids can
contain up to 3 IDs.
examples:
- - character_01hx8p0demo
duration:
type: string
enum:
- '4'
- '6'
- '8'
- '10'
description: >-
The duration of the generated video in seconds.
Available values are 4, 6, 8, and 10. When video input
is provided, the output duration is determined by the
model automatically. This duration parameter will not
take effect.Note: when video input is provided, the
output duration is determined by the model
automatically. This duration parameter will not take
effect.
examples:
- '8'
aspect_ratio:
type: string
enum:
- '16:9'
- '9:16'
description: >-
The aspect ratio of the generated video. `16:9` is
landscape, and `9:16` is portrait.
examples:
- '16:9'
seed:
type: integer
description: >-
Random seed. Range: [0, 2147483647]. If not specified,
the system generates a seed automatically. Fixing the
seed can improve reproducibility, but results may still
vary due to the model’s stochasticity.
resolution:
type: string
enum:
- 720p
- 1080p
- 4k
description: >-
The resolution of the generated video. Available values
are 720p, 1080p, and 4k.
default: 720p
examples:
- 720p
x-apidog-orders:
- prompt
- image_urls
- audio_ids
- video_list
- character_ids
- duration
- aspect_ratio
- resolution
- seed
required:
- prompt
- duration
x-apidog-orders:
- model
- callBackUrl
- input
example:
model: gemini-omni-video
callBackUrl: https://your-domain.com/api/callback
input:
prompt: >-
Create a futuristic night city short film with a slow push-in
shot as the character walks out from a neon-lit street.
image_urls:
- https://example.com/assets/scene-1.png
- https://example.com/assets/scene-2.png
audio_ids:
- audio_01hx8p0demo
video_list:
- url: https://example.com/assets/source-video.mp4
start: 0
ends: 10
duration: '4'
responses:
'200':
description: Request successful
content:
application/json:
schema:
allOf:
- type: object
properties: {}
- type: object
properties:
data:
type: object
properties:
taskId:
type: string
description: >-
Task ID, which can be used to query task status
through the task detail endpoint.
examples:
- task_gemini_1765180586443
x-apidog-orders:
- taskId
x-apidog-orders:
- data
example:
code: 200
msg: success
data:
taskId: task_gemini_1765180586443
headers: {}
x-apidog-name: ''
security:
- BearerAuth: []
x-apidog:
schemeGroups:
- id: kn8M4YUlc5i0A0179ezwx
schemeIds:
- BearerAuth
required: true
use:
id: kn8M4YUlc5i0A0179ezwx
scopes:
kn8M4YUlc5i0A0179ezwx:
BearerAuth: []
x-apidog-folder: docs/en/Market/Video Models/Gemini Omni
x-apidog-status: released
x-run-in-apidog: https://app.apidog.com/web/project/1184766/apis/api-36212673-run
components:
schemas: {}
securitySchemes:
BearerAuth:
type: bearer
scheme: bearer
bearerFormat: API Key
description: >-
All API requests require a Bearer Token. Add the header `Authorization:
Bearer YOUR_API_KEY` to authenticate requests.
BearerAuth1:
type: bearer
scheme: bearer
bearerFormat: API Key
description: >-
所有 API 请求都需要 Bearer Token。请在请求头中添加 `Authorization: Bearer YOUR_API_KEY`
进行身份验证。
servers:
- url: https://api.kie.ai
description: 正式环境
security:
- BearerAuth: []
x-apidog:
schemeGroups:
- id: kn8M4YUlc5i0A0179ezwx
schemeIds:
- BearerAuth
required: true
use:
id: kn8M4YUlc5i0A0179ezwx
scopes:
kn8M4YUlc5i0A0179ezwx:
BearerAuth: []
```