Models/google/veo-3-1-fast

Veo 3.1 Fast

prospolabs/veo-3-1-fast

Google Veo 3.1 Fast tier — balanced speed and quality with best-in-class lip sync when audio is enabled. Up to 1080p, 8 seconds max.

GoogleImage to VideoCommercial usetext-to-videoimage-to-videoaudio
Up to 40% off

Starts from

$0.060/ second

Retail $0.100

Pay-as-you-go — no subscription, no tokens.

Input

0 chars
$0.090/s$0.150/s40% offper clip

Billed by output length — final charge = rate × clip seconds.

Typically completes in 3090s.

Result

Idle
View API →
Try a sample video

Or write your own prompt and hit Generate.

Pricing: $0.06/sec audio off, $0.09/sec audio on — 40% off market. See full pricing →

01overview

What is Veo 3.1 Fast?

Veo 3.1 Fast is Google's speed-tuned tier of Veo 3.1: it generates video roughly twice as fast as the standard model while holding nearly the same visual quality, with best-in-class lip sync when audio is enabled. On Prospolabs you can run it from $0.06/sec audio-off (retail $0.10/sec) or $0.09/sec audio-on (retail $0.15/sec) — about 40% under list price.

It does text-to-video and image-to-video (supply a starting frame), outputs up to 1080p in 16:9 or 9:16, and produces clips of 4, 6, or 8 seconds. Audio is optional, so you only pay the higher rate when you actually want synchronized sound and dialogue.

Prospolabs is pay-per-generation in plain USD — no tokens, no monthly seat, no subscription. The price is the same whether you generate in the UI or over the API, failed runs are auto-refunded, and you can top up from $5 and start rendering.

Cheapest on the web

Veo 3.1 Fast from $0.06/sec — about 40% under retail

Prospolabs runs Veo 3.1 Fast at $0.06/sec audio-off (retail $0.10/sec) and $0.09/sec audio-on (retail $0.15/sec) — roughly 40% off either way. A 5-second clip is $0.30 audio-off or $0.45 audio-on; an 8-second clip is $0.48 or $0.72. No tokens, no subscription — pay-per-generation in USD, the same price in the UI and the API, with auto-refunds on failed runs. Top up from $5 and start rendering.

02capabilities

What Veo 3.1 Fast can do

01

Text-to-video

Describe a scene, action, and camera move in natural language and Veo 3.1 Fast returns a coherent clip with natural motion. Use the text mode, set duration (4s/6s/8s), resolution (720p/1080p), and aspect_ratio (16:9 or 9:16).

02

Image-to-video

Supply a single starting frame and the model extrapolates motion and depth while staying faithful to the source image. Use the image mode to animate stills, product shots, or a generated keyframe into an 8-second clip.

03

Optional audio with best-in-class lip sync

Turn audio on for synchronized dialogue, ambient sound, and effects generated natively with the video — lip sync is best-in-class on this tier. Audio is optional on Prospolabs, so audio-off runs stay at the cheaper $0.06/sec rate.

04

Controls for repeatable output

Tune duration, resolution, and aspect ratio, steer with negative_prompt to suppress unwanted elements, and pin a seed for reproducible results across iterations and A/B drafts.

05

Fast iteration tier

Veo 3.1 Fast is built for speed — close to the standard model's look at a fraction of the cost — so it suits draft passes, social and product clips, and high-volume batches where turnaround matters.

03use cases

What people build with Veo 3.1 Fast

Social and short-form video

Spin up 9:16 vertical clips for reels and shorts in 4-8 second beats, then iterate cheaply on Prospolabs at $0.06/sec audio-off before committing to a final pass with audio on.

Product and e-commerce motion

Animate a single product still with image-to-video to add motion, depth, and camera movement for listings, ads, and landing pages — an 8-second audio-off clip lands at $0.48.

Talking-head and dialogue clips

Enable audio for synchronized speech with best-in-class lip sync — well suited to explainers, ads, and character lines. A 5-second audio-on clip is $0.45, an 8-second clip $0.72.

Rapid drafts and A/B testing

Use a fixed seed and negative_prompt to render variations quickly, validate a concept on the Fast tier, and only step up to a higher-fidelity model for the hero deliverable.

04api

Call Veo 3.1 Fast from your code

The Prospolabs API is async: submit a job, poll for status, then download. POST to /v1/generate with model id "veo-3-1-fast" and a "mode" field ("text" or "image"). The API and UI cost the same USD rate, and failed runs are auto-refunded.

Full Veo 3.1 Fast API reference →
  1. 01

    POST /v1/generate with model "veo-3-1-fast", your mode ("text" for text-to-video, "image" with a starting frame for image-to-video), and prompt.

  2. 02

    Set controls: duration (4s/6s/8s, max 8s), resolution (720p/1080p, default 1080p), aspect_ratio (16:9 or 9:16), plus optional negative_prompt and seed.

  3. 03

    Toggle audio: leave it off to bill at $0.06/sec, or turn it on for synchronized dialogue and best-in-class lip sync at $0.09/sec.

  4. 04

    Read the generation_id from the response and poll for status until the job completes.

  5. 05

    Download from output_url (valid 7 days). If a run fails, Prospolabs auto-refunds it — see /models/veo-3-1-fast/api for full reference.

?questions

Veo 3.1 Fast FAQ

  • It is priced per second: $0.06/sec with audio off (retail $0.10/sec) and $0.09/sec with audio on (retail $0.15/sec) — about 40% under list. So a 5-second clip is $0.30 or $0.45, and an 8-second clip is $0.48 or $0.72. Prospolabs bills the same USD rate in the UI and the API.

  • Yes. On Prospolabs audio is optional: audio-off runs are billed at $0.06/sec and audio-on runs (which add synchronized sound and best-in-class lip sync) at $0.09/sec. Leave audio off for silent clips to pay the lower rate.

  • Fast is Google's speed-tuned tier — it generates roughly twice as fast as standard Veo 3.1 while holding nearly the same look, at a fraction of the cost. Use Fast for drafts, social clips, and high-volume work; step up to /models/veo-3-1 when you want maximum fidelity.

  • Up to 1080p (default 1080p, 720p also available), in 16:9 or 9:16, with clip lengths of 4, 6, or 8 seconds — 8 seconds is the maximum per generation.

  • Both. Set mode "text" for text-to-video, or mode "image" and supply a starting frame for image-to-video. Both run on the same Veo 3.1 Fast model id and the same per-second pricing on Prospolabs.

  • No. Prospolabs is pay-per-generation in USD — no tokens and no monthly subscription. You top up from $5, pay only for what you render, and failed runs are auto-refunded.

  • POST to /v1/generate with model id "veo-3-1-fast" and a mode field, then poll the returned generation_id and download from output_url (valid 7 days). Full details are at /models/veo-3-1-fast/api.

related