API reference

Videos

Creates a video generation with one of 11 models. Options differ by model — see Models or GET /models. When the video starts from a character or a prompt, Orior first generates a starting frame (included in the price as one image).

The request returns 202 with a queued generation. Videos typically take 1–6 minutes.

Create a video

POST/v1/videos

Scope: generateIdempotency-Key requiredReturns 202

Provide at least one of prompt, character_id, character_image, start_frame or (Motion Control) motion_video.

Body

modelstringrequired
seedance-2.5, seedance-2.0, seedance-2.0-mini, kling-3.0, kling-3.0-motion-control, wan-2.6, minimax-h3, gemini-omni, grok-imagine-1.5, veo-3, veo-3-fast.
promptstring
Motion, action and camera direction. Maximum length depends on the model (5,000–20,000).
character_idstring
Character to feature; a first frame is generated from it.
character_imagestring (media)
A face to use without saving a character. Upload ID (upl_…), public https:// URL, or base64 data URL.
product_idstring
Product to feature (combine with character_id for product-in-hand).
start_framestring (media)
Exact first frame. Upload ID (upl_…), public https:// URL, or base64 data URL.
reference_imagesarray
Extra tagged references to mention in the prompt (e.g. “wearing the jacket from @ref1”). Up to 8; each model receives up to its own limit, the rest shape the first frame.
- tagstring
  Tag used in the prompt, @ + up to 20 letters/digits/underscores. Defaults to @ref1, @ref2, …
- imagestring (media)required
  Upload ID (upl_…), public https:// URL, or base64 data URL.
narration_audiostring (media)
Voice-over/audio reference for models that use it. Upload ID (upl_…), public https:// URL, or base64 data URL.
durationstring
Seconds, as allowed by the model (e.g. "8").
resolutionstring
As allowed by the model (e.g. 720p, 1080p, 2K, 4k). For kling-3.0: 720p = standard, 1080p = pro.
aspect_ratiostring
As allowed by the model. adaptive follows the first frame.
audioboolean
Generate sound (models with audio support).
motion_videostring (media)
kling-3.0-motion-control only (required): a 1–30 s clip whose movement is transferred. Priced per measured second.
motion_orientationstringdefault: video
kling-3.0-motion-control only: follow the orientation of the video or the image.

Errors

Code	When
`VALIDATION_ERROR`	Unknown model, option not offered by the model, prompt too long, or motion_video missing/too long.
`INSUFFICIENT_CREDITS`	Balance below the price.
`CONCURRENCY_LIMIT`	Too many API jobs running.

Request

curl -X POST https://api.oriorai.com/v1/videos \
  -H "Authorization: Bearer $ORIOR_API_KEY" \
  -H "Idempotency-Key: $(uuidgen)" \
  -H "Content-Type: application/json" \
  -d '{
  "model": "seedance-2.5",
  "prompt": "She turns to camera and laughs, handheld",
  "character_id": "Qx8p2LmN4vRt",
  "duration": "8",
  "resolution": "720p",
  "aspect_ratio": "9:16",
  "audio": true
}'

const response = await fetch('https://api.oriorai.com/v1/videos', {
  method: 'POST',
  headers: {
    Authorization: `Bearer ${process.env.ORIOR_API_KEY}`,
    'Idempotency-Key': crypto.randomUUID(),
    'Content-Type': 'application/json',
  },
  body: JSON.stringify({
    "model": "seedance-2.5",
    "prompt": "She turns to camera and laughs, handheld",
    "character_id": "Qx8p2LmN4vRt",
    "duration": "8",
    "resolution": "720p",
    "aspect_ratio": "9:16",
    "audio": true
  }),
})
const data = await response.json()
if (!response.ok) throw new Error(`${data.error.code}: ${data.error.message}`)

import os, uuid
import requests

response = requests.post(
    "https://api.oriorai.com/v1/videos",
    headers={"Authorization": f"Bearer {os.environ['ORIOR_API_KEY']}", "Idempotency-Key": str(uuid.uuid4())},
    json={
        "model": "seedance-2.5",
        "prompt": "She turns to camera and laughs, handheld",
        "character_id": "Qx8p2LmN4vRt",
        "duration": "8",
        "resolution": "720p",
        "aspect_ratio": "9:16",
        "audio": True,
    },
    timeout=60,
)
response.raise_for_status()
data = response.json()

Response · 202

Response

{
  "id": "gen_7f3c9a1e2b4d4c8f9a0b1c2d3e4f5a6b",
  "object": "generation",
  "type": "video",
  "model": "seedance-2.5",
  "status": "queued",
  "credits": 46,
  "refunded_credits": 0,
  "result": null,
  "error": null,
  "request": {
    "model": "seedance-2.5",
    "prompt": "She turns to camera and laughs, handheld",
    "character_id": "Qx8p2LmN4vRt",
    "product_id": null,
    "duration": "8",
    "resolution": "720p",
    "aspect_ratio": "9:16",
    "audio": true,
    "has_start_frame": false,
    "reference_images": 0,
    "motion_video_seconds": null
  },
  "timeline": [],
  "created_at": 1758620000000,
  "completed_at": null
}