Cheaper Gen.

Video models

AI video generation API

22 official AI video models — Seedance, Kling, Veo, Gemini Omni, Wan, Hailuo and more — through one REST API. From $0.0176/s, up to 82% below the official list price, billed per second of output.

Models
22
10 developers
From
$0.0176/s
PixVerse V6
Below official
up to 82%
Grok Imagine Video 1.5
Failed runs
$0
refunded automatically

Live prices

Every AI video model, compared

Starting price per second at each model's lowest tier, next to the official list price. Select a column to sort.

AI video generation API: models, specs and prices. Select a column header to sort.
T2V · I2V · Ref2V
4K
4–15 s
Included
$0.0569/s
480p
$0.0703/s
up to 20%
T2V · I2V · Ref2V
1080p
4–15 s
Included
$0.0877/s
480p
$0.1075/s
up to 18%
T2V · I2V · Ref2V
720p
4–15 s
Included
$0.0438/s
480p
$0.0562/s
up to 22%
T2V · I2V · Ref2V
720p
4–15 s
Included
$0.02/s
480p
$0.0352/s
up to 43%
T2V · I2V
4K
3–15 s
Optional (+)
$0.0739/s
720p
$0.084/s
up to 13%
T2V · I2V · Ref2V
4K
3–15 s
Optional (+)
$0.0739/s
720p
$0.084/s
up to 12%
T2V · I2V
1080p
3–15 s
—
$0.099/s
720p
$0.112/s
up to 12%
T2V · I2V
4K
4, 6, 8 s
Optional (+)
$0.1719/s
720p + audio
$0.40/s
up to 57%
T2V · I2V
4K
4, 6, 8 s
Optional (+)
$0.0413/s
720p + audio
$0.10/s
up to 59%
T2V · I2V
720p
4, 6, 8 s
—
$0.0206/s
720p
$0.05/s
up to 59%
T2V · I2V · Ref2V
720p
4, 6, 8, 10 s
—
$0.0722/s
720p · 8 s
$0.1017/s
up to 29%
T2V · I2V · Ref2V
1080p
2–30 s
Included
$0.0362/s
480p
$0.05/s
up to 28%
T2V · I2V · Ref2V
1080p
2–30 s
Included
$0.0566/s
480p
$0.068/s
up to 19%
T2V · I2V · Ref2V
2K
5–15 s
—
$0.044/s
768P
$0.08/s
up to 45%
T2V · I2V
768P
5–15 s
—
$0.0415/s
480P
$0.05/s
up to 21%
T2V · I2V
1080p
1–16 s
Included
$0.0332/s
540p
$0.035/s
up to 5%
Ref2V
1080p
3–15 s
—
$0.11/s
720p
$0.12/s
up to 9%
I2V · Ref2V
4K
3–16 s
Included
$0.0396/s
540p
$0.045/s
up to 12%
T2V · Ref2V
1080p
1–15 s
Included
$0.0248/s
720p
$0.14/s
up to 82%
T2V · I2V
720p
5–20 s
Included
$0.1425/s
720p
$0.15/s
up to 5%
T2V · I2V · Ref2V
1080p
1–15 s
Optional (+)
$0.0176/s
360p
$0.022/s
up to 20%
T2V · I2V
1080p
1–8 s
Included
$0.022/s
720p
$0.09/s
up to 76%

Prices are live from our catalog. Official list prices are the developers' own published API prices. Audio “Optional (+)” means clips with audio are billed at a higher rate. Every tier: full price list.

How to choose

Picking an AI video model

Lowest price
PixVerse V6 from $0.0176/s (360p).
Highest resolution
Seedance 2.0, Kling 3.0, Kling 3.0 Omni, Veo 3.1 and 2 more — up to 4K.
Longest clips
Wan 3.0 and Wan 3.0 Prime — up to 30 s per generation.
Audio included
Seedance 2.0, Seedance 2.5, Seedance 2.0 Fast, Seedance 2.0 Mini and 7 more generate synchronized sound at no extra cost.
Most references
Seedance 2.5 — up to 30 reference images and 10 reference videos.
Most aspect ratios
PixVerse V6 — 8 aspect ratios.
Biggest discount
Grok Imagine Video 1.5: up to 82% below xAI's official price.

How it works

About the AI Video API

One API for every video model

Submit a job with POST /v1/run/{endpoint} and get a generation id back; poll it or pass a webhook_url. Text-to-video, image-to-video and reference-to-video all work the same way, so switching models means changing the endpoint and its input fields — not your integration.

Billed per second

Video is charged per second of generated output at the resolution you pick. The price is held when you submit and refunded in full if the generation fails.

Head to head

Popular comparisons

All comparisons

FAQ

AI Video API questions

An AI video API lets your code generate video clips from a text prompt, a still image or reference media. On Cheaper Gen. you call 22 official models — including Seedance 2.0, Seedance 2.5, Seedance 2.0 Fast, Seedance 2.0 Mini, Kling 3.0 and 17 more — through one endpoint shape and one key.

PixVerse V6 starts at $0.0176/s at 360p, so a 5-second clip costs $0.088. The cheap AI video API page ranks every model by price.

Per second of output, at the resolution you choose. Some models include audio in the price; others bill audio at a higher rate — the table shows which. Failed generations are free.

There is no subscription. Credits are prepaid from $10 and work across every model in the API and the Studio; failed generations are refunded automatically, and the quote endpoint prices any request for free before you run it.

No. One Cheaper Gen. API key and balance covers ByteDance, Kuaishou, Google, Alibaba, MiniMax, Shengshu, xAI, Black Forest Labs, PixVerse and Lightricks. Every model uses the same request shape: POST /v1/run/{endpoint}, then poll or receive a webhook.