> ## Documentation Index
> Fetch the complete documentation index at: https://docs.easy-peasy.ai/llms.txt
> Use this file to discover all available pages before exploring further.
# Generate Video
> Generate AI videos from text prompts or images using 40+ models including Veo 3.1, Kling 3.0, Sora 2, Hailuo, and more.
Video generation is asynchronous — the response returns a video ID immediately. Use the [Get Video](/api-reference/endpoint/get-video) endpoint to poll for the result.
**Input types:**
- **Text-to-video** — Provide only a `prompt`
- **Image-to-video** — Provide both `prompt` and `image` URL
**Note:** Video generation requires a paid plan.
## Workflow
Video generation is asynchronous. Here's the typical flow:
1. **Submit** a video generation request using this endpoint
2. **Save** the `id` from the response
3. **Poll** the [Get Video](/api-reference/endpoint/get-video) endpoint every 15–30 seconds until `status` is `completed`
4. **Download** the video from the `url` field
Video generation typically takes **1–5 minutes** depending on the model, duration, and resolution. A paid plan is required.
## Input types
The `model` value must **exactly match** one of the titles in the `model` enum, and each mode is a **separate model variant** — the title's suffix tells you which image inputs it accepts:
* **Text-to-video** — Provide only a `prompt` (no image fields). Use the base title, e.g. `Seedance 2.0`, `Veo 3.1`.
* **Image-to-video (first frame)** — Provide `prompt` + `image` (a start-frame URL). Use a title ending in **"Image"**, e.g. `Seedance 2.0 Image`, `Veo 3.1 Fast Image`, `Kling 3.0 Pro Image`.
* **First & last frame** — Provide `prompt` + `image` (first) + `tailImage` (last). Use a **"First-Last Frame"** title (e.g. `Veo 3.1 First-Last Frame`), or `Seedance 2.0 Image`, which also accepts a `tailImage` end frame.
* **Reference-to-video (multiple references)** — Provide `prompt` + `referenceImages` (an array of `{ "url": ... }`, up to 9) to keep characters/objects/style consistent. Use a title ending in **"Reference"**, e.g. `Seedance 2.0 Reference`, `Kling O1 Reference`, `Kling O3 Reference`.
For Seedance 2.0 these are distinct models: `Seedance 2.0` (text), `Seedance 2.0 Image` (first frame, optional last frame), and `Seedance 2.0 Reference` (up to 9 reference images). A single request cannot combine a start frame **and** reference images — pick the variant that matches your input.
Image URLs for `image`, `tailImage`, and `referenceImages[].url` do not need to be hosted on Easy-Peasy, but they **must be downloadable server-side by our video provider**. Some hosts (e.g. Wikimedia, or sites that block hotlinking / non-browser requests) return **HTTP 403** to automated downloads — these fail with a `Could not download the input (HTTP 403)` error. If in doubt, host the image on your own public CDN/bucket or on Easy-Peasy.
### First frame (image-to-video)
```bash theme={null}
curl -X POST https://easy-peasy.ai/api/generate-video \
-H "x-api-key: YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"prompt": "The subject slowly turns toward the camera, cinematic lighting",
"model": "Seedance 2.0 Image",
"image": "https://yourcdn.com/first-frame.png",
"duration": "5",
"aspectRatio": "16:9",
"resolution": "1080p"
}'
```
Add `"tailImage": "https://yourcdn.com/last-frame.png"` to interpolate between a first and last frame.
### Multiple reference images (reference-to-video)
```bash theme={null}
curl -X POST https://easy-peasy.ai/api/generate-video \
-H "x-api-key: YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"prompt": "The character walks through a neon-lit market at night",
"model": "Seedance 2.0 Reference",
"referenceImages": [
{ "url": "https://yourcdn.com/character.png" },
{ "url": "https://yourcdn.com/outfit.png" },
{ "url": "https://yourcdn.com/environment.png" }
],
"duration": "5",
"aspectRatio": "16:9",
"resolution": "1080p"
}'
```
`referenceImages[].type` (`character` / `object` / `style`) is optional and only used by the Kling O1/O3 Reference models to separate character/object references from style references. Seedance 2.0 Reference treats every entry as a general reference, so you can send just `{ "url": ... }`.
## Polling example
```javascript theme={null}
async function waitForVideo(videoId, apiKey) {
const maxAttempts = 40; // ~10 minutes max
for (let i = 0; i < maxAttempts; i++) {
const response = await fetch(
`https://easy-peasy.ai/api/get-video?video_id=${videoId}`,
{
headers: { 'x-api-key': apiKey },
}
);
const data = await response.json();
if (data.video.status === 'completed') {
return data.video.url; // Video URL is ready!
}
await new Promise(resolve => setTimeout(resolve, 15000));
}
throw new Error('Video generation timeout');
}
```
## OpenAPI
````yaml POST /api/generate-video
openapi: 3.0.1
info:
title: Easy-Peasy.AI API
description: >-
OpenAPI Specifications for the Easy-Peasy.AI API.
All API requests must be authenticated with an API key. Include the
`x-api-key` API key in the request header with all requests. You can get the
API key [here](https://easy-peasy.ai/settings/api).
version: 1.0.4
servers:
- url: https://easy-peasy.ai
security:
- apiKeyAuth: []
paths:
/api/generate-video:
post:
summary: Generate Video
description: >-
Generate AI videos from text prompts or images using 40+ models
including Veo 3.1, Kling 3.0, Sora 2, Hailuo, and more.
Video generation is asynchronous — the response returns a video ID
immediately. Use the [Get Video](/api-reference/endpoint/get-video)
endpoint to poll for the result.
**Input types:**
- **Text-to-video** — Provide only a `prompt`
- **Image-to-video** — Provide both `prompt` and `image` URL
**Note:** Video generation requires a paid plan.
operationId: generateVideo
parameters:
- name: x-api-key
in: header
required: true
schema:
type: string
description: Your API key
example: 27feb2bb-aeb4-4a83-9fb6-8f3f2a15885e
requestBody:
required: true
content:
application/json:
schema:
$ref: '#/components/schemas/GenerateVideoRequest'
examples:
text_to_video:
summary: Text-to-video
value:
prompt: A cat walking on the beach at sunset, cinematic lighting
model: Veo 3.1 Fast
duration: '5'
aspectRatio: '16:9'
image_to_video:
summary: Image-to-video
value:
prompt: The subject slowly turns and smiles at the camera
image: https://example.com/photo.jpg
model: Kling 3.0 Pro Image
duration: '5'
aspectRatio: '16:9'
responses:
'200':
description: Video generation started
content:
application/json:
schema:
$ref: '#/components/schemas/GenerateVideoResponse'
example:
id: 12345
prompt: A cat walking on the beach at sunset
image_url: ''
model: Veo 3.1 Fast
is_video: true
created_at: '2025-01-15T10:30:00.000Z'
'400':
description: Bad request — invalid model, duration, aspect ratio, or input type
content:
application/json:
schema:
$ref: '#/components/schemas/Error'
example:
error: Invalid model selected
'401':
description: Unauthorized
content:
application/json:
schema:
$ref: '#/components/schemas/Error'
example:
error: Invalid API key
'403':
description: Plan limit reached or free plan
content:
application/json:
schema:
$ref: '#/components/schemas/Error'
example:
error: >-
Video generation is not available on the Free plan. Please
upgrade to a paid plan to generate videos.
components:
schemas:
GenerateVideoRequest:
type: object
required:
- prompt
- model
properties:
prompt:
type: string
description: Text prompt describing the video to generate.
example: A cat walking on the beach at sunset, cinematic lighting
image:
type: string
format: uri
description: >-
URL of a start/first-frame image for image-to-video generation. Use
a model whose title ends in "Image" (e.g. "Seedance 2.0 Image", "Veo
3.1 Image"). If omitted, text-to-video mode is used. Any publicly
reachable URL is accepted.
model:
type: string
description: >-
The video model to use (required). The value must exactly match one
of the enum titles below. Titles ending in "Image" take a
start-frame `image`; titles ending in "Reference" take
`referenceImages`; "First-Last Frame" titles take both `image` and
`tailImage`; all others are text-to-video.
enum:
- Veo 3.1 Fast
- Grok Imagine
- Grok Imagine Reference
- Seedance 2.0 Fast
- Seedance 2.0 Turbo
- Seedance 2.0
- Seedance 2.0 Mini
- Seedance 2.0 Reference
- Seedance 2.0 Fast Reference
- Seedance 2.0 Mini Reference
- Kling 2.6 Pro
- Veo 3.1 Lite
- Veo 3.1 Lite Image
- Veo 3.1 Fast Image
- Grok Imagine Image
- Grok Imagine 1.5
- Seedance 2.0 Image
- Seedance 2.0 Fast Image
- Seedance 2.0 Turbo Image
- Seedance 2.0 Mini Image
- Happy Horse
- Happy Horse Image
- Happy Horse Reference
- Kling 2.6 Pro Image
- Kling 3.0 Pro
- Kling 3.0 Pro Image
- Kling 3.0 Standard
- Kling 3.0 Standard Image
- Kling O3 Pro
- Kling O3 Pro Image
- Kling O3 Standard
- Kling O3 Standard Image
- Kling O3 Reference
- Seedance 1.5 Pro
- Seedance 1.5 Pro Image
- Kling Motion Control Pro
- Kling Motion Control
- Kling O1 Image
- Kling O1 Reference
- Kling 2.5 Turbo Pro Image
- Kling 2.5 Turbo Pro
- Kling 2.5 Turbo Standard
- Seedance 1.5 Pro First-Last Frame
- Veo 3.1
- LTX-2 Pro
- LTX-2 Fast
- Veo 3.1 Image
- Veo 3.1 First-Last Frame
- Veo 3.1 Fast First-Last Frame
- LTX-2 Pro Image
- LTX-2 Fast Image
- Sora 2
- Sora 2 Pro
- Hailuo 2.0
- Hailuo 2.0 Pro
- Hailuo 2.3
- Hailuo 2.3 Pro
- Hailuo 2.3 Image
- Hailuo 2.3 Pro Image
- Hailuo 2.3-Fast Pro Image
- Seedance v1 Pro Fast
- Pixverse v5.5
- Wan 2.5
- Wan v2.2 Turbo
- Wan v2.2
- Sora 2 Image
- Sora 2 Pro Image
- Hailuo 2.0 Image
- Hailuo 2.0 Pro Image
- Seedance 1.0 Pro Fast
- Pixverse v5.5 Image
- Wan 2.5 Image
- Wan v2.2 Turbo Image
- Wan-2.2 Image
duration:
type: string
default: '5'
description: >-
Duration of the video in seconds. Available values depend on the
model.
example: '5'
aspectRatio:
type: string
default: '16:9'
description: Aspect ratio of the video.
enum:
- '16:9'
- '9:16'
- '1:1'
- '4:3'
- '3:4'
- '21:9'
- '9:21'
- '3:2'
- '2:3'
resolution:
type: string
description: >-
Video resolution. Allowed values depend on the model (e.g. Seedance
2.0 supports 480p–4k; Veo 3.1 supports 720p/1080p).
enum:
- 480p
- 720p
- 1080p
- 4k
generateAudio:
type: boolean
description: >-
Whether to generate audio along with the video (supported by some
models like Veo 3.1, Kling 3.0, Seedance 2.0).
tailImage:
type: string
format: uri
description: >-
URL of an end/last-frame image. Only used by models that support a
tail frame — the "First-Last Frame" titles and "Seedance 2.0 Image"
(paired with `image` as the first frame). Any publicly reachable URL
is accepted.
referenceImages:
type: array
description: >-
Reference images for reference-to-video models (titles ending in
"Reference", e.g. "Seedance 2.0 Reference", "Kling O1 Reference",
"Kling O3 Reference"). Used to keep characters/objects/style
consistent across the generated video. Seedance 2.0 Reference
accepts up to 9 images. Not used by text-to-video or image-to-video
models.
items:
type: object
required:
- url
properties:
url:
type: string
format: uri
description: Publicly reachable image URL.
type:
type: string
enum:
- character
- object
- style
description: >-
Reference role. Used by Kling O1/O3 to distinguish
character/object references (as `elements`) from style
references. Seedance 2.0 treats every entry as a general
reference, so `type` is optional for it.
GenerateVideoResponse:
type: object
properties:
id:
type: integer
description: >-
Video ID. Use this to poll for the result with the Get Video
endpoint.
prompt:
type: string
description: The prompt used for generation
image_url:
type: string
description: Video URL. Empty string while processing.
model:
type: string
description: The model used for generation
is_video:
type: boolean
created_at:
type: string
format: date-time
Error:
type: object
properties:
error:
type: string
description: Error message
example: Invalid API key
securitySchemes:
apiKeyAuth:
type: apiKey
in: header
name: x-api-key
description: >-
API key for authentication. Get yours at
https://easy-peasy.ai/settings/api
````