OkayPic API Documentation

Generate AI videos and images programmatically via our REST API.

Authentication

All API requests require a Bearer token. Create an API key in your Developer Options page.

Authorization: Bearer okpk_your_api_key_here

Available Models

Video Models

Model ID	Label	Pricing	Duration	Ratios
jimeng-video-seedance-2.0	Seedance 2.0	$0.12/s ($0.24/s with video ref)	4–15s	16:9, 9:16, 1:1, 4:3, 3:4, 3:2, 2:3, 21:9
jimeng-video-seedance-2.0-fast	Seedance 2.0 Fast	$0.08/s ($0.16/s with video ref)	4–15s	16:9, 9:16, 1:1, 4:3, 3:4, 3:2, 2:3, 21:9
grok-video-3	Grok Video 3	$0.02/s	5/10/15s	2:3, 3:2, 1:1
veo3.1	Veo 3.1	See variants below	—	16:9, 9:16

Veo 3.1 Variants

Variant ID	Label	Cost	Max Images
veo3.1-fast	Fast	$0.10	2
veo3.1-components	Components	$0.15	3
veo3.1-pro	Pro	$0.30	2

Image Models

Model ID	Label	Cost	Ratios	Resolutions
jimeng-5.0	Jimeng 5.0	$0.20	1:1, 16:9, 9:16, 4:3, 3:4, 3:2, 2:3, 21:9	1k, 2k, 4k

Endpoints

POST/api/video/generate

Submit a video generation task.

Request Body

Parameter	Type	Required	Description
model	string	Yes	Video model ID (see models table)
prompt	string	Yes	Text description of the video to generate
duration	number	No	Video duration in seconds (model-dependent range)
ratio	string	No	Aspect ratio (e.g. "16:9", "9:16", "1:1")
variant	string	No	For Veo 3.1 only: variant ID (see variants table)
ref_mode	string	No	Seedance only: `"omni"` or `"first-last"`. Required when media is provided. See below.
media	array	No	Reference media array. Each item: {type: "image"\|"video"\|"audio", url: "..."}. Constraints depend on model and ref_mode.

Seedance Reference Modes

Seedance 2.0 / 2.0 Fast require ref_mode when media is provided.

ref_mode	Description	Media Constraints
"omni"	Omni Reference — images, video, or audio	0–9 items total, max 1 video. Types: image, video, audio
"first-last"	First & Last Frame — image-only keyframes	0–2 images only. First item = first frame, second = last frame

Cost note: using a video as reference doubles the per-second rate.

Other Models — Media Reference

Grok Video 3: 0–1 reference image. No ref_mode needed.
Veo 3.1: 0–3 reference images (depends on variant, see table). No ref_mode needed.

Response

{
  "generationId": "uuid",
  "taskId": "backend-task-id",
  "status": "pending"
}

GET/api/video/tasks/{taskId}

Poll a task for its current status and results.

Response (completed)

{
  "status": "completed",
  "resultUrls": ["https://cdn.example.com/video.mp4"],
  "type": "video"
}

Response (processing)

{
  "status": "processing",
  "progress": 50,
  "type": "video"
}

Response (failed)

{
  "status": "failed",
  "errorMessage": "Generation failed",
  "type": "video"
}

Code Examples

# 1. Submit a video generation task (text-only)
curl -X POST https://okaypic.com/api/video/generate \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "jimeng-video-seedance-2.0",
    "prompt": "A cat playing piano in a jazz club",
    "duration": 5,
    "ratio": "16:9"
  }'

# With reference images (omni mode — images/video/audio, up to 9)
curl -X POST https://okaypic.com/api/video/generate \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "jimeng-video-seedance-2.0",
    "prompt": "Make this cat dance",
    "duration": 5,
    "ratio": "16:9",
    "ref_mode": "omni",
    "media": [{"type": "image", "url": "https://example.com/cat.jpg"}]
  }'

# With first & last frames (image-only, 0–2 images)
curl -X POST https://okaypic.com/api/video/generate \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "jimeng-video-seedance-2.0",
    "prompt": "Smooth transition",
    "duration": 5,
    "ratio": "16:9",
    "ref_mode": "first-last",
    "media": [
      {"type": "image", "url": "https://example.com/frame1.jpg"},
      {"type": "image", "url": "https://example.com/frame2.jpg"}
    ]
  }'

# 2. Poll for task completion
curl https://okaypic.com/api/video/tasks/TASK_ID \
  -H "Authorization: Bearer YOUR_API_KEY"

# Response: { "status": "completed", "resultUrls": ["https://..."], "type": "video" }

Error Codes

Status	Meaning
400	Bad request — invalid model, missing parameters, or non-video model
401	Unauthorized — invalid or missing API key
402	Insufficient balance
429	Too many concurrent tasks (max 3)
500	Internal server error — backend generation failed