curl --request POST \
  --url https://api.evolink.ai/v1/videos/generations \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "model": "gemini-omni-1.2-flash-text-to-video",
  "prompt": "A glass marble rolls quickly along a wooden track, then falls into water with a splash.",
  "quality": "720p",
  "duration": 3,
  "aspect_ratio": "16:9"
}
'

{
  "created": 1757169743,
  "id": "task-unified-1757169743-7cvnl5zw",
  "model": "gemini-omni-1.2-flash-text-to-video",
  "object": "video.generation.task",
  "progress": 0,
  "status": "processing",
  "task_info": {
    "can_cancel": false,
    "estimated_time": 60,
    "video_duration": 3
  },
  "type": "video",
  "usage": {
    "billing_rule": "per_token",
    "credits_reserved": 102.34,
    "user_group": "default"
  }
}

{
  "error": {
    "code": "invalid_request",
    "message": "Invalid request parameters",
    "type": "invalid_request_error"
  }
}

{
  "error": {
    "code": "unauthorized",
    "message": "Invalid or expired token",
    "type": "authentication_error"
  }
}

{
  "error": {
    "code": "insufficient_quota",
    "message": "Insufficient quota. Please top up your account.",
    "type": "insufficient_quota"
  }
}

{
  "error": {
    "code": "model_access_denied",
    "message": "Token does not have access to model: gemini-omni-1.2-flash-text-to-video",
    "type": "invalid_request_error"
  }
}

{
  "error": {
    "code": "rate_limit_exceeded",
    "message": "Too many requests, please try again later",
    "type": "rate_limit_error"
  }
}

{
  "error": {
    "code": "internal_error",
    "message": "Internal server error",
    "type": "api_error"
  }
}

Gemini Omni 1.2 Flash

Gemini Omni 1.2 Flash Full Parameter Overview

Gemini Omni 1.2 Flash provides six video task types through POST /v1/videos/generations.
Text-to-video: gemini-omni-1.2-flash-text-to-video; requires only a text prompt; 3–15 seconds.
Image-to-video: gemini-omni-1.2-flash-image-to-video; one first-frame image (3–15 seconds) or two first/last-frame images (3–10 seconds).
Reference-to-video: gemini-omni-1.2-flash-reference-to-video; reference images and up to 3 reference videos. When duration exceeds 10 seconds, at most 1 reference video is allowed.
Video editing: gemini-omni-1.2-flash-video-edit; one input video of up to 15 seconds and optionally up to 10 reference images. quality defaults to auto.
Video extension: gemini-omni-1.2-flash-video-extend; one input video of up to 30 seconds and optionally up to 10 reference images. quality defaults to auto; duration is the number of seconds to append.
Video upscaling: gemini-omni-1.2-flash-video-upscale; one video of up to 15 seconds in a supported standard size. Specify the target resolution with quality; no prompt is needed.
Aspect ratio: Defaults to auto. Editing, extension, and upscaling preserve the input video’s aspect ratio.
The model generates synchronized native audio automatically; no extra switch is needed.
Set seed to improve reproducibility.
Asynchronous processing: query task status using the returned task ID.
Generated video links remain valid for 24 hours; save them promptly.

POST

videos

generations

curl --request POST \
  --url https://api.evolink.ai/v1/videos/generations \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "model": "gemini-omni-1.2-flash-text-to-video",
  "prompt": "A glass marble rolls quickly along a wooden track, then falls into water with a splash.",
  "quality": "720p",
  "duration": 3,
  "aspect_ratio": "16:9"
}
'

{
  "created": 1757169743,
  "id": "task-unified-1757169743-7cvnl5zw",
  "model": "gemini-omni-1.2-flash-text-to-video",
  "object": "video.generation.task",
  "progress": 0,
  "status": "processing",
  "task_info": {
    "can_cancel": false,
    "estimated_time": 60,
    "video_duration": 3
  },
  "type": "video",
  "usage": {
    "billing_rule": "per_token",
    "credits_reserved": 102.34,
    "user_group": "default"
  }
}

{
  "error": {
    "code": "invalid_request",
    "message": "Invalid request parameters",
    "type": "invalid_request_error"
  }
}

{
  "error": {
    "code": "unauthorized",
    "message": "Invalid or expired token",
    "type": "authentication_error"
  }
}

{
  "error": {
    "code": "insufficient_quota",
    "message": "Insufficient quota. Please top up your account.",
    "type": "insufficient_quota"
  }
}

{
  "error": {
    "code": "model_access_denied",
    "message": "Token does not have access to model: gemini-omni-1.2-flash-text-to-video",
    "type": "invalid_request_error"
  }
}

{
  "error": {
    "code": "rate_limit_exceeded",
    "message": "Too many requests, please try again later",
    "type": "rate_limit_error"
  }
}

{
  "error": {
    "code": "internal_error",
    "message": "Internal server error",
    "type": "api_error"
  }
}

Authorizations

Authorization

string

header

required

##All endpoints require authentication using a Bearer Token##

Get your API Key:

Visit the API Key management page to obtain your API Key

Add it to the request header:

Authorization: Bearer YOUR_API_KEY

Body

application/json

model

enum<string>

required

Model name. Select the model ID that matches the input and task:

gemini-omni-1.2-flash-text-to-video: text-to-video
gemini-omni-1.2-flash-image-to-video: image-to-video (first frame or first/last frames)
gemini-omni-1.2-flash-reference-to-video: reference-to-video
gemini-omni-1.2-flash-video-edit: video editing
gemini-omni-1.2-flash-video-extend: video extension
gemini-omni-1.2-flash-video-upscale: video upscaling

Available options:

gemini-omni-1.2-flash-text-to-video,

gemini-omni-1.2-flash-image-to-video,

gemini-omni-1.2-flash-reference-to-video,

gemini-omni-1.2-flash-video-edit,

gemini-omni-1.2-flash-video-extend,

gemini-omni-1.2-flash-video-upscale

Example:

"gemini-omni-1.2-flash-text-to-video"

prompt

string

Text prompt describing the generation, editing, or continuation goal; supports Chinese and English. Required for every mode except video upscaling, which needs no prompt.

Describe the subject, action, scene, camera movement, and elements to preserve clearly.
Put negative requirements directly in the prompt, such as No dialogue or No text. There is no separate negative prompt parameter.

Video upscaling does not use prompt; any supplied prompt is ignored.

Example:

"A glass marble rolls quickly along a wooden track, then falls into water with a splash."

image_urls

string<uri>[]

Image input array.

Image-to-video: required; 1–2 images. One is the first frame; two are the first and last frames.
Reference-to-video: required; the published recommendation is at most 10 reference images.
Video editing: optional; up to 10 reference images.
Video extension: optional; up to 10 reference images.
Text-to-video and video upscaling: image input is not supported.

HTTP/HTTPS image URLs only; supported formats: png, jpeg, webp.

Required array length: 1 - 10 elements

Example:

[
  "https://example.com/character.png",
  "https://example.com/scene.png"
]

video_urls

string<uri>[]

Video input array.

Reference-to-video: optional; the published recommendation is at most 3 videos. When duration exceeds 10 seconds, at most 1 is allowed.
Video editing: exactly 1 required, up to 15 seconds.
Video extension: exactly 1 required, up to 30 seconds.
Video upscaling: exactly 1 required, up to 15 seconds. Supported sizes: 640×360, 1280×720, 1920×1080, and their portrait equivalents.
Text-to-video and image-to-video: video input is not supported.

Each reference video must be no longer than 10 seconds.

Required array length: 1 - 3 elements

Example:

["https://example.com/reference.mp4"]

quality

enum<string>

Video resolution.

Options (landscape / portrait):

360p: 640×360 / 360×640
720p: 1280×720 / 720×1280
1080p: 1920×1080 / 1080×1920
4k: 3840×2160 / 2160×3840
auto: editing and extension only; follows the input resolution.

Defaults:

Text-to-video, image-to-video, reference-to-video: 720p.
Video editing and extension: auto.
Video upscaling: required target resolution, higher than the input. 360p → 720p or 1080p; 720p → 1080p or 4k; 1080p → 4k.

Billing: Text-to-video, image-to-video, reference-to-video, editing, and extension are billed separately for input tokens, video output tokens, and other output tokens (including thinking). Video upscaling is billed only for video output tokens, with no input or thinking token charges. The charge at submission is an estimate; final billing uses actual usage.

Available options:

360p,

720p,

1080p,

4k,

auto

Example:

"720p"

duration

Video duration.

Text-to-video and reference-to-video: integer 3–15 seconds; default 3.
Image-to-video: 3–15 seconds with one first frame, or 3–10 seconds with first/last frames; default 3.
Video editing and upscaling: only auto, which is also the default. Output duration follows the input.
Video extension: integer 3–10 seconds; default 3. This is the duration to append.

Required range: 3 <= x <= 15

Example:

3

aspect_ratio

enum<string>

default:auto

Video aspect ratio; default auto.

16:9: landscape.
9:16: portrait.
auto: the model selects the ratio from the prompt and input; this is the default.

Editing, extension, and upscaling allow only auto and preserve the input video's aspect ratio.

Available options:

16:9,

9:16,

auto

Example:

"auto"

seed

integer

Random seed; random by default.

Range: 0–4294967295.
Fixing the seed reduces variation from parameter changes when tuning prompts and improves reproducibility.

Required range: 0 <= x <= 4294967295

Example:

42

callback_url

string<uri>

HTTPS callback URL to notify when the task completes

Callback timing:

Triggered when the task completes (completed), fails (failed), or is cancelled (cancelled)
Sent after billing is confirmed

Security restrictions:

HTTPS protocol only
Callbacks to internal IP addresses are forbidden (127.0.0.1, 10.x.x.x, 172.16-31.x.x, 192.168.x.x, etc.)
URL length must not exceed 2048 characters

Callback mechanism:

Timeout: 10 seconds
Up to 3 retries on failure (retried 1s / 2s / 4s after each failure respectively)
The callback body format matches the response of the task query endpoint
A 2xx status code from the callback URL is treated as success; other status codes trigger a retry

Example:

"https://your-domain.com/webhooks/video-task-completed"

Response

Video task created successfully

created

integer

Task creation timestamp

Example:

1757169743

string

Task ID

Example:

"task-unified-1757169743-7cvnl5zw"

model

string

The model name actually used

Example:

"gemini-omni-1.2-flash-text-to-video"

object

enum<string>

The specific type of the task

Available options:

video.generation.task

progress

integer

Task progress percentage (0-100)

Required range: 0 <= x <= 100

Example:

0

status

enum<string>

Task status. The create endpoint always returns processing; the internal pending state is also exposed to clients as processing.

Available options:

pending,

processing,

completed,

failed

Example:

"processing"

task_info

object

Detailed video task information

Show child attributes

type

enum<string>

The output type of the task

Available options:

text,

image,

audio,

video

Example:

"video"

usage

object

Usage and billing information

Show child attributes

Image Series

Video Series

Audio Series

Text Series

Account Management

Task Management

File Management

Gemini Omni 1.2 Flash Full Parameter Overview

Authorizations

Body

Response