vidu-video

Vidu Q3 Video Generation

Vidu's latest Q3 Pro text-to-video model supports multiple resolutions (360p–1080p), customizable duration and aspect ratio, and can directly generate videos with audio.

Available Models

Model ID	Feature	Description
fal-ai/vidu/q3/text-to-video	Text-to-Video	Generate video from a text description, with support for audio output

Pricing

Billed by video seconds × resolution :

Resolution	Price (credits/second)	5-second video	10-second video
360p / 540p	28 credits/second	140 credits	280 credits
720p / 1080p	62 credits/second	310 credits	620 credits

Workflow

1. Call submit_task

Use the MCP tool submit_task to submit a task:

{
  "model_id": "fal-ai/vidu/q3/text-to-video",
  "parameters": {
    "prompt": "In an ultra-realistic fashion photography style featuring light blue and pale amber tones, an astronaut in a spacesuit walks through the fog.",
    "duration": 5,
    "resolution": "720p",
    "aspect_ratio": "16:9",
    "audio": true
  }
}

Parameter Description

Parameter	Type	Required	Default Value	Description
prompt	string	Yes	-	Video generation prompt (max 2000 characters)
duration	integer	No	5	Video duration (seconds), 1-16
resolution	string	No	"720p"	Resolution: 360p / 540p / 720p / 1080p
aspect_ratio	string	No	"16:9"	Aspect ratio: 16:9 / 9:16 / 4:3 / 3:4 / 1:1
audio	boolean	No	true	Whether to generate video with audio
seed	integer	No	-	Random seed, used to reproduce results

Query Task Status

After submitting a task, task_idis returned; use get_task to query the result:

{
  "task_id": "返回的任务ID"
}

Task status:

• pending - Queued
• processing - Processing
• completed - Completed, result is in result in
• failed - Failed, check the error field

Complete Example

User request: Generate a video of an astronaut walking in fog

Execution Steps：

1. Call submit_task：

{
  "model_id": "fal-ai/vidu/q3/text-to-video",
  "parameters": {
    "prompt": "In an ultra-realistic fashion photography style featuring light blue and pale amber tones, an astronaut in a spacesuit walks through the fog.",
    "duration": 5,
    "resolution": "720p",
    "aspect_ratio": "16:9"
  }
}

1. Get task_id then call get_task to query the result

Prompt Tips

1. Describe clearly: Describe the scene, action, lighting, and color tone in detail
2. English first: Using English prompts usually works better
3. Style keywords: Add style words such as "cinematic", "ultra-realistic", "photography style"
4. Resolution selection: 360p/540p are low-cost and suitable for testing; 720p/1080p are suitable for final output
5. Audio feature: audio=true is enabled by default, and the generated video will include sound

Notes

• 720p and 1080p resolutions are priced at 2.2 times that of 360p/540p
• Video duration supports flexible selection from 1-16 seconds
• Supports multiple aspect ratios, suitable for content needs on different platforms