Veo 3.1 Models

Overview

Vendor: Google. The Veo 3.1 family supports text, image, frame, and reference workflows. veo-3-1 emphasizes frame and multi-image reference control; Fast prioritizes speed, and Lite has a fixed configured per-generation price.

Supported modes

Each model supports text-to-video, image-to-video, frames-to-video, and reference-to-video.

Model	Durations	Resolutions	Aspect ratios	Audio
`veo-3-1`	No duration option is configured	720p, 1080p, 4k	16:9, 9:16, Auto	No
`veo-3-1-fast`	No duration option is configured	720p, 1080p, 4k	16:9, 9:16, Auto	No
`veo-3-1-lite`	No duration option is configured	720p, 1080p, 4k	16:9, 9:16, Auto	No

For frames-to-video, provide at least two image_urls. For reference-to-video, provide one to three image_urls.

Credit pricing

Credits are per generation and are the same across the four supported modes.

Model	720p	1080p	4k
`veo-3-1`	225	240	285
`veo-3-1-fast`	30	40	100
`veo-3-1-lite`	15	35	75

cURL

curl -X POST https://vidrush-ai.com/api/v1/videos \
  -H "Authorization: Bearer $VIDRUSH_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "veo-3-1-fast",
    "mode": "text-to-video",
    "prompt": "A glass terrarium on a sunlit desk",
    "options": { "resolution": "720p", "aspect_ratio": "16:9" }
  }'

CLI

vidrush generate -m veo-3-1-fast -p "A glass terrarium on a sunlit desk" \
  --resolution 720p --aspect-ratio 16:9 --wait

MCP

Ask the MCP client to use generate_video with a Veo model and omit duration, because this family exposes no duration option.

Notes

audio: true is invalid for these configurations.
Auto is a valid aspect_ratio value for every Veo 3.1 mode.