> ## Documentation Index
> Fetch the complete documentation index at: https://docs.dimilinks.com/llms.txt
> Use this file to discover all available pages before exploring further.
# Create video generation
> Generate 6- to 20-second videos with grok-imagine-video from text or reference images.
`/v1/videos/generations` returns a `task_id` immediately after submission. Generation time depends on duration and resolution, and is generally one to five minutes. Poll [`/v1/tasks/{task_id}`](/en/api-reference/videos/grok-imagine-video/tasks) for status.
The video endpoint currently supports only the `grok-imagine-video` model. Inference and billing use the special pricing for the Grok route.
## Request URL
```http theme={null}
POST https://dimilinks.com/v1/videos/generations
Content-Type: application/json
Authorization: Bearer
```
## Text-to-video example
```bash theme={null}
curl "https://dimilinks.com/v1/videos/generations" \
-H "Authorization: Bearer $DIMILINKS_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "grok-imagine-video",
"prompt": "A group of neon jellyfish drifting slowly through the deep sea in cinematic light",
"seconds": "6",
"size": "1280x720",
"resolution_name": "720p",
"preset": "normal"
}'
```
Successful response:
```json theme={null}
{
"data": {
"id": "video_xxx",
"task_id": "video_xxx",
"status": "queued"
}
}
```
After receiving `task_id`, call [Retrieve video task](/en/api-reference/videos/grok-imagine-video/tasks).
## Reference-image-to-video example
Convert reference images to `data:image/...;base64,...` Data URLs first. Browser `blob:` URLs cannot be submitted directly; your server must read the binary data and encode it.
```bash theme={null}
curl "https://dimilinks.com/v1/videos/generations" \
-H "Authorization: Bearer $DIMILINKS_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "grok-imagine-video",
"prompt": "Keep the person consistent while the camera makes one complete orbit around the subject",
"seconds": "10",
"size": "1280x720",
"resolution_name": "720p",
"preset": "normal",
"input_reference": [
"data:image/png;base64,iVBORw0KGgo..."
]
}'
```
## Parameters
| Parameter | Type | Required | Description |
| ----------------- | --------- | -------- | ------------------------------------------------------------------------------------------------------------------- |
| `model` | string | Yes | Currently fixed to `grok-imagine-video`. |
| `prompt` | string | Yes | Video description, up to 2,500 characters. |
| `seconds` | string | Yes | Duration in seconds: `"6"` / `"10"` / `"12"` / `"16"` / `"20"`. |
| `size` | string | Yes | Frame size: `1280x720` (16:9), `720x1280` (9:16), or `1024x1024` (1:1). |
| `resolution_name` | string | Yes | Resolution tier: `720p` (HD) or `480p` (SD). |
| `preset` | string | Yes | Style preset: `normal` (standard), `fun` (high playfulness), `spicy` (high intensity), or `custom` (custom prompt). |
| `input_reference` | string\[] | No | Array of up to seven reference-image Data URLs. When provided, enables reference-image-to-video mode. |
| `group` | string | No | Route group. Console requests include `default`; third-party callers generally leave it empty. |
## Duration and resolution combinations
* Use `720p` for final output. `480p` generates faster and is suitable for previews.
* Longer durations increase generation time and price. Start with `seconds: "6"` to validate the workflow.
* When changing the same prompt to `9:16` or `1:1`, describe where the subject should appear in the frame to avoid cropping.
## Error handling
The video endpoint also uses the OpenAI-style `error` wrapper. See [Errors](/en/api-reference/errors) for common errors. If a `400` response reports a reference-image problem, verify that:
* The reference image is PNG, JPEG, or WebP.
* It uses a Data URL, not a `blob:` URL or regular HTTP URL.
* Each reference image is smaller than 10 MB.
* The request includes no more than seven reference images.
## OpenAPI
````yaml POST /videos/generations
openapi: 3.1.0
info:
title: DimiLinks API
description: >-
DimiLinks unified API for chat (OpenAI-compatible and Anthropic native),
image generation/editing, and video generation.
version: 1.1.0
servers:
- url: https://dimilinks.com/v1
description: OpenAI 兼容入口：chat completions / images / videos / models / tasks
security:
- bearerAuth: []
tags:
- name: Models
- name: Chat
- name: Images
- name: Videos
- name: Tasks
paths:
/videos/generations:
post:
tags:
- Videos
summary: Create video generation
operationId: createVideoGeneration
requestBody:
required: true
content:
application/json:
schema:
$ref: '#/components/schemas/VideoGenerationRequest'
responses:
'200':
description: Video task submission
content:
application/json:
schema:
$ref: '#/components/schemas/VideoSubmitResponse'
'400':
$ref: '#/components/responses/BadRequest'
'401':
$ref: '#/components/responses/Unauthorized'
'403':
$ref: '#/components/responses/Forbidden'
'429':
$ref: '#/components/responses/RateLimited'
components:
schemas:
VideoGenerationRequest:
type: object
properties:
model:
type: string
default: grok-imagine-video
prompt:
type: string
maxLength: 2500
seconds:
type: string
enum:
- '6'
- '10'
- '12'
- '16'
- '20'
size:
type: string
enum:
- 1280x720
- 720x1280
- 1024x1024
resolution_name:
type: string
enum:
- 480p
- 720p
preset:
type: string
enum:
- normal
- fun
- spicy
- custom
input_reference:
type: array
items:
type: string
description: data:image/...;base64,... Data URL
maxItems: 7
group:
type: string
required:
- model
- prompt
- seconds
- size
- resolution_name
- preset
VideoSubmitResponse:
type: object
properties:
data:
type: object
properties:
id:
type: string
example: video_xxx
task_id:
type: string
example: video_xxx
status:
type: string
enum:
- queued
- pending
- submitted
ErrorResponse:
type: object
properties:
error:
$ref: '#/components/schemas/ErrorObject'
ErrorObject:
type: object
properties:
message:
type: string
type:
type: string
example: invalid_request_error
code:
oneOf:
- type: string
- type: integer
responses:
BadRequest:
description: Invalid request
content:
application/json:
schema:
$ref: '#/components/schemas/ErrorResponse'
Unauthorized:
description: Missing or invalid API key
content:
application/json:
schema:
$ref: '#/components/schemas/ErrorResponse'
Forbidden:
description: Model is not allowed for this API key
content:
application/json:
schema:
$ref: '#/components/schemas/ErrorResponse'
RateLimited:
description: Rate limit exceeded
content:
application/json:
schema:
$ref: '#/components/schemas/ErrorResponse'
securitySchemes:
bearerAuth:
type: http
scheme: bearer
````