Veo 3.1 on Sume: the audio-on rate equals Gemini API's 720p list

Sume's Veo 3.1 audio-on list rates equal the 720p lines on Google's Gemini API pricing page, and Sume bills those list rates times 1.25: $0.50, $0.125 and $0.0625 per second for Standard, Fast and Lite. The Gemini API page lists Standard at $0.40 for 720p and 1080p, Fast at $0.10 for 720p, and Lite at $0.05 for 720p (Gemini API pricing, read 2026-10-10).

The extra 25 percent is Sume's house margin on every video row, as the Video Generation docs say: Sume reserves the provider list times 1.25 at submit.

Where Sume's numbers come from

Sume's catalog stores Veo rates from the Vertex pricing page dated 2026-10-08, with a separate row for video plus audio and one for video only. The code comment says posted Cloud Billing line items, not these, are what the Google credit pays, so treat the numbers as list rates.

What Sume does not offer from that page

The Gemini API page also lists 1080p and 4K tiers: Standard 4K at $0.60, Fast 1080p at $0.12 and 4K at $0.30, Lite 1080p at $0.08. Sume's Veo rows run at 720p only, so those tiers are not available here. If you need 1080p or 4K on Sume, Gemini Omni Flash 1.1 lists 360p, 720p, 1080p and 4K.

Two checks that save money

Audio changes the Sume bill on every Veo row, so set it deliberately:

• generate_audio defaults to true. Send false for a silent clip at the lower rate: a Standard 8-second clip is $4.00 with audio and $2.00 without.
• Lite 8 seconds with audio is $0.50; Fast 8 seconds with audio is $1.00.

Which tier fits

Pick Lite for a cheap 720p check of a prompt, Fast when you iterate, Standard for the final. All three accept 4, 6 or 8 seconds in 16:9 or 9:16, text prompts only.