Generate videos from text prompts or images using 6 models across 2 providers.

Endpoints

Method	Endpoint	Description
`POST`	`/api/v1/video/generate`	Text-to-video or image-to-video
`GET`	`/api/v1/video/status`	Poll async video jobs

MiniMax Hailuo

MiniMax Hailuo video generation is available via the existing Media API. The Video API documented here covers additional providers (Google Veo, Kling, Wan).

Available Models

Model	ID	Provider	Type	Notes
Veo 3.1	`veo-3.1`	Google	Async	720p/1080p/4K, native audio, 4-8s
Veo 3.1 Fast	`veo-3.1-fast`	Google	Async	Faster, slightly lower quality
Kling 3 Pro	`kling-3-pro`	fal.ai	Sync	High quality, text-to-video
Kling 3 Pro (i2v)	`kling-3-pro-i2v`	fal.ai	Sync	Image-to-video
Wan 2.6	`wan-2.6`	fal.ai	Sync	Text-to-video
Wan 2.6 (i2v)	`wan-2.6-i2v`	fal.ai	Sync	Image-to-video

Async models (Google Veo) return an operation_id immediately — poll /api/v1/video/status for results. Sync models (fal.ai) block until the video is ready (can take 1-3 minutes).

Auto Image-to-Video

If you pass image_url with kling-3-pro or wan-2.6, the API automatically switches to the image-to-video variant. No need to specify the -i2v suffix manually.

Generate Video

POST /api/v1/video/generate

Request

{
  "model": "veo-3.1",
  "prompt": "A cinematic sunset over ocean waves, slow motion",
  "duration": "8",
  "aspect_ratio": "16:9",
  "resolution": "1080p"
}

Parameters

Parameter	Type	Required	Description
`prompt`	string	Yes	Video description
`model`	string	No	Model ID (default: `veo-3.1`)
`image_url`	string	No	Source image for image-to-video
`duration`	string	No	Duration: `"4"`, `"6"`, `"8"` seconds (Google Veo)
`aspect_ratio`	string	No	`"16:9"` or `"9:16"`
`resolution`	string	No	`"720p"`, `"1080p"`, `"4k"` (Google Veo)

Response (Async — Google Veo)

{
  "success": true,
  "data": {
    "model": "veo-3.1",
    "provider": "google",
    "status": "processing",
    "operation_id": "operations/abc123",
    "poll_url": "/api/v1/video/status?provider=google&operation_id=operations%2Fabc123"
  }
}

Response (Sync — fal.ai)

{
  "success": true,
  "data": {
    "model": "kling-3-pro",
    "provider": "fal",
    "status": "completed",
    "video_url": "https://fal-cdn.example.com/video.mp4"
  }
}

Poll Video Status

GET /api/v1/video/status?provider=google&operation_id=operations/abc123

For async models (Google Veo), poll this endpoint until done: true.

Parameters

Parameter	Type	Required	Description
`provider`	string	Yes	`"google"`
`operation_id`	string	Yes	Operation ID from generate response

Response (Processing)

{
  "success": true,
  "data": {
    "provider": "google",
    "operation_id": "operations/abc123",
    "done": false
  }
}

Response (Complete)

{
  "success": true,
  "data": {
    "provider": "google",
    "operation_id": "operations/abc123",
    "done": true,
    "video_url": "gs://storage-uri/video.mp4"
  }
}

Polling Interval

Poll every 10 seconds. Video generation typically takes 1-3 minutes depending on model, duration, and resolution.

Examples

# 1. Submit video generation
curl https://concurred.ai/api/v1/video/generate \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "veo-3.1",
    "prompt": "A drone shot over a misty mountain forest at sunrise",
    "duration": "8",
    "resolution": "1080p"
  }'
 
# 2. Poll until done (use operation_id from response)
curl "https://concurred.ai/api/v1/video/status?provider=google&operation_id=operations/abc123" \
  -H "Authorization: Bearer YOUR_API_KEY"

Video API

On this page