Veo 3.1: how many Lite takes cost one Standard take (5 at 1080p)

At 1080p, one Veo 3.1 Standard take costs the same as five Lite takes or about 3.3 Fast takes. At 720p, one Standard take equals eight Lite takes or four Fast takes. Those ratios come straight from Google's per-second prices and tell you how much better Standard has to be to pay for itself.

They hold for any clip length, because every tier is billed per second.

The ratios

Divide Standard's price by the cheaper tier's price at the same resolution. At 1080p: $0.40 / $0.08 = 5 for Lite, and $0.40 / $0.12 = 3.33 for Fast. At 720p: $0.40 / $0.05 = 8 for Lite and $0.40 / $0.10 = 4 for Fast. At 1080p, Fast against Lite is $0.12 / $0.08 = 1.5.

Lite has no 4K, so the only 4K ratio is Standard to Fast: $0.60 / $0.30 = 2.

How to use the ratio

If you expect to need four Fast takes to get a keeper at 720p and one Standard take to get the same, the two cost the same. If Fast needs three and Standard one, Fast is cheaper. If Fast needs five, Standard is cheaper.

At 4K the ratio is only 2, so Standard is a more reasonable bet there than at 720p, where it must beat Fast four to one.

The caveat

We do not know your keep rates, and neither does the pricing page. Count them with ten test prompts at each tier, then use the ratios. A tier that looks better in one demo reel may not look better on your product.

Also remember that cost is not the only limit. Which tier you can use also depends on the clip lengths and inputs Google allows for it, which are on Google's Veo page and which we did not read.

Sume equivalent

Sume does not list a Veo model. Its Google video model is gemini-omni-flash-1.1, which takes 3 to 10 seconds at 360p, 720p, 1080p or 4K, in 16:9 or 9:16, with audio always on. Sume bills the provider list rate times 1.25 for each output second, so its rates are $0.0375, $0.125, $0.1875 and $0.375 a second. Its resolution steps give ratios as well: one 4K second equals three 720p seconds or two 1080p seconds, and one 720p second equals 3.33 360p seconds.