Guides

How much does one AI video cost? Veo, Kling, SeeDance and Grok per clip (2026)

· 5 min read

Get the iPhone app

Dora is made by Saga Labs, the publisher of this site.

Short answer: in Dora, a short AI video clip (4 to 10 seconds) of the models below costs between about $0.14 and $3.28, depending on the model, the length, the resolution and whether it has sound. Grok Imagine and SeeDance 1.5 Pro are the cheapest, Kling 2.6 and Veo 3.1 sit in the middle, and SeeDance 2.0 at 720p is the most expensive. Most apps hide this behind credits; below, every price is turned into dollars per clip.

Cost per clip, in dollars

Prices Dora charges on October 25, 2026. One coin costs about one US cent: the 1,000-coin pack is $9.99 (the monthly plan works out a little cheaper per coin).

Model (maker) Clip Coins in Dora About
Grok Imagine (xAI) 6 s, 480p 15 $0.15
Grok Imagine (xAI) 10 s, 720p 45 $0.45
SeeDance 1.5 Pro (ByteDance), with sound 4 s, 480p 14 $0.14
SeeDance 1.5 Pro (ByteDance), with sound 8 s, 1080p 120 $1.20
Kling 2.6 (Kuaishou), no sound 5 s 55 $0.55
Kling 2.6 (Kuaishou), with sound 10 s 220 $2.20
Veo 3.1 (Google) one clip 60 $0.60
SeeDance 2.0 Fast (ByteDance) 4 s, 480p 62 $0.62
SeeDance 2.0 (ByteDance) 8 s, 720p 328 $3.28

What other apps and APIs charge

  • Kling's own app: a 5-second 720p clip without sound uses 30 credits, and top-up credits cost $1 for 66, according to CloudZero. That is about $0.45 per clip.
  • invideo: Veo 3 uses 4 to 12.8 of its credits per clip depending on length and resolution (invideo help); the dollar value depends on your plan.
  • APIs (for developers): Veo 3.1 Fast is listed at about $0.10 per second at 720p and SeeDance 2.0 Fast at about $0.09 per second, but third-party price lists disagree with each other (GMI Cloud, Atlas Cloud).

Why the same model costs different amounts

  1. Length: most models charge per second, so 8 seconds costs twice as much as 4.
  2. Resolution: higher resolution costs more on the same model. In Dora, 4 seconds of SeeDance 1.5 Pro is 14 coins at 480p and 60 at 1080p; 6 seconds of Grok Imagine is 15 at 480p and 27 at 720p.
  3. Sound: on some models it costs extra. In Dora, Kling 2.6 doubles (55 coins for 5 seconds silent, 110 with sound) and Kling 3.0 at 720p goes from 14 to 18 coins per second; SeeDance 1.5 Pro costs the same with or without sound.
  4. Retries: the real cost is per clip you keep. If a model refuses a photo or the result is wrong, you pay again. In Dora, coins come back automatically when a generation fails.

How to spend less per video

  • Draft at 480p and 4 seconds; make only the final version at 720p or 1080p.
  • Test the prompt on a cheap model (Grok Imagine, SeeDance 1.5 Pro), then use the expensive one.
  • Leave sound off if you will add music in TikTok or CapCut.
  • Check the price on the button before you generate: every app shows it, in credits or coins.

Dora has all of these models in one account, on the App Store and on the web (pricing).

Frequently asked questions

What is the cheapest AI video model?

In Dora, SeeDance 1.5 Pro and Grok Imagine: 14 coins (about $0.14) for 4 seconds of SeeDance 1.5 Pro at 480p, 15 coins (about $0.15) for 6 seconds of Grok Imagine at 480p.

How much does a Veo 3.1 video cost?

In Dora, 60 coins (about $0.60) per clip. For developers, third-party API price lists put Veo 3.1 Fast at about $0.10 per second at 720p.

Why do apps use credits instead of dollars?

Because each model costs a different amount per second, and credits let one plan cover all of them. Divide the plan price by its credits to get the dollar value of one credit.

Do I pay if the video fails?

It depends on the app. In Dora, coins come back automatically when a generation fails.

Sources

Get the iPhone appUse it in Claude or ChatGPT

Keep reading

ComparisonsThe best AI video generators in 2026The AI video generators worth your time in 2026, ranked, and the easiest way to use all of them under one plan.ModelsVeo 3.1, explainedGoogle's flagship text-to-video — prompt fidelity, realistic motion, and keyframe transitions.ModelsSeeDance 2, explainedByteDance's cinematic model — image-to-video, character consistency and audio.