APIMaster.ai

Veo 3.1 video generation API

Use Google official model IDs through one asynchronous API. Official and Ext provider model names are mapped on the server; submission, polling and download stay the same.

Veo 3.1 video generation API

Use Google official model IDs through one asynchronous API. Official and Ext provider model names are mapped on the server; submission, polling and download stay the same.

  • veo-3.1-generate-preview — Veo 3.1
  • veo-3.1-fast-generate-preview — Veo 3.1 Fast

Parameters and route capabilities

Defaults: 8 seconds, 720p, 16:9, audio enabled. Official supports 4, 6 or 8 seconds; Ext requires 8. Both 1080p and 4K require 8 seconds. One video per task.

Official supports generate_audio:false, negative_prompt, seed (including 0), person_generation, resize_mode and enhance_prompt:true. Ext cannot disable audio. Official-only parameters can fall back to an available official route. Do not use internal provider model IDs.

image_urls accepts a first frame or first/last frames. Fast Ext also accepts three reference images (generation_type:reference, 16:9 only). Standard has no three-image reference mode. Use publicly accessible HTTP(S) image URLs.

{"model":"veo-3.1-generate-preview","prompt":"A paper boat drifting across a quiet lake at sunrise, natural ambient sound","duration":8,"resolution":"720p","aspect_ratio":"16:9","generate_audio":true}
curl "https://apimaster.ai/v1/videos" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"veo-3.1-generate-preview","prompt":"A paper boat drifting across a quiet lake at sunrise","duration":8,"resolution":"720p","generate_audio":true}'

POST https://apimaster.ai/v1/videos → {"id":"task_...","task_id":"task_...","status":"queued"}

An id acknowledges submission, not completion. Poll every 5–10 seconds until completed or failed. Download promptly with the same API key after completion.

curl "https://apimaster.ai/v1/videos/task_..." -H "Authorization: Bearer YOUR_API_KEY"
curl -L "https://apimaster.ai/v1/videos/task_.../content" -H "Authorization: Bearer YOUR_API_KEY" -o video.mp4

queued → in_progress → completed / failed

Billing and refunds

The table lists Google official base USD/second prices. Route coefficients and account pricing determine the final charge; check model cards and usage logs. Each Ext call is billed as 8 seconds. An extension is another 8-second billing unit, not the combined original-plus-extension length. Failed tasks refund the generation reservation through asynchronous settlement.

Google model 720p + audio 1080p + audio 4K + audio 720p silent 1080p silent 4K silent
veo-3.1-generate-preview $0.40/s $0.40/s $0.60/s $0.20/s $0.20/s $0.40/s
veo-3.1-fast-generate-preview $0.10/s $0.12/s $0.30/s $0.08/s $0.10/s $0.25/s

Google Gemini API · Google Cloud

Extend a video

Verified on 2026-10-17: extensions work on Standard Ext (veo-3.1-generate-preview). Fast extensions are currently rejected upstream. Generate the original video with Standard if you plan to extend it.

Only completed Ext tasks can be extended, using the same model, owner and original route. Official tasks cannot use this extension endpoint. raw:true requests only the extension. Invalid parameters return 400; a pinned route without the requested capability returns 503.

curl "https://apimaster.ai/v1/videos/task_.../remix" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"veo-3.1-generate-preview","prompt":"The paper boat continues past green reeds","raw":false,"resolution":"720p","aspect_ratio":"16:9"}'