Veo 3.1 draft ladder: five Lite takes then one Standard final, $5.20

Five 8-second Veo 3.1 Lite drafts at 720p plus one Standard final at 1080p cost $5.20 in total: $2.00 for the drafts and $3.20 for the final. Six Standard takes at 1080p would cost $19.20, so the ladder costs about 27% as much.

The ladder only helps if a Lite draft predicts what the Standard render will do, and that is a bet you should test on your own prompts.

The arithmetic

Lite is $0.05 a second at 720p, so an 8-second draft is $0.40. Standard is $0.40 a second at both 720p and 1080p, so an 8-second final is $3.20. Fast at 1080p is $0.12 a second, or $0.96 for 8 seconds.

A ladder spends cheaply on the unknowns, which are framing, motion and prompt wording, and expensively once.

The last row is the sensible one

If Fast at 1080p is good enough for the final, the ladder ends at $2.96. Standard is worth its price only when you can see what it adds. A useful test is to render the same approved draft at Fast and Standard once, compare, and decide for the batch.

Standard costs the same at 720p and 1080p, so there is no saving in rendering a Standard draft at 720p.

Where it fails

Drafts at a cheaper tier are different samples, not previews. A good Lite take is not a promise that Standard produces the same shot from the same prompt, and Google's pricing page does not say otherwise. If you need the exact framing you saw, use a first frame or reference image where the model supports it.

On Sume, Veo is not listed. The nearest pattern is Omni Flash: a 360p draft at $0.0375 a second and a 1080p final at $0.1875, which for a 10-second clip is $0.375 and $1.875.

A cheaper version of the ladder

If Lite and Standard look too far apart for a draft to predict the final, use Fast at 720p as the draft tier. Five 8-second Fast drafts at $0.10 a second cost $4.00, and a Standard 1080p final is $3.20, so $7.20 in all. That is still well under six Standard takes, and the draft is closer to the final tier.

The reverse is also worth testing: if a Lite draft is nearly as good as Fast, you may never need the Fast tier for the batch.

Before you build it

Count how many drafts you really keep. If most of your drafts are rejected for prompt errors, fix the prompt first and spend less on every tier.