Veo 3.1 Lite (Preview) on Sume: what the preview label means
Sume's catalog names the row 'Veo 3.1 Lite (Preview)'. Its limits, the audio rates, and how it differs from the Gemini API's Lite lines for 720p and 1080p.

Sume lists the cheapest Veo row as veo-3.1-lite with the display name "Veo 3.1 Lite (Preview)"; it runs text-to-video at 720p for 4, 6 or 8 seconds, in 16:9 or 9:16, and bills $0.0625 per second with audio or $0.0375 without. The catalog code carries the Preview label in the name only; it does not add a different rule set.
Google's Gemini API pricing page lists Veo 3.1 Lite at $0.05 per second for 720p and $0.08 for 1080p (read 2026-10-10); Sume has the 720p tier only.
What the row accepts
The three Veo rows share one request builder, so Lite has the same envelope as Standard and Fast. The builder accepts a prompt, duration of 4, 6 or 8 (default 8), resolution 720p, aspect_ratio 16:9 (default) or 9:16, and a boolean generate_audio (default true). It refuses image, frame, reference and video inputs, bitrate_mode and model_params.
| Item | Value |
|---|---|
| Id | veo-3.1-lite |
| Name | Veo 3.1 Lite (Preview) |
| Durations | 4, 6, 8 s |
| Resolution | 720p |
| Aspect ratios | 16:9, 9:16 |
| Billed with audio | $0.0625 per second |
| Billed without audio | $0.0375 per second |
| Route | Google Cloud (Vertex) only |
What Preview means for you
A preview label on the vendor side usually means limits can change. Sume's row name carries that warning, so do not hard-code assumptions: read supported_durations, supported_resolutions and pricing_skus from GET /v1/videos/models at run time, and expect the row to change or disappear if Google changes it.
Cost of ten drafts
Ten 4-second silent drafts on Lite cost 10 x 4 x $0.0375 = $1.50. With audio the same ten clips cost 10 x 4 x $0.0625 = $2.50. That is a cheap way to test a prompt in the Veo family before you spend on Standard, where the same ten silent 4-second clips cost 10 x 4 x $0.25 = $10.00.
When Lite is the wrong pick
Choose a different row when the job needs a start image, 1080p or more, or more than 8 seconds. Gemini Omni Flash 1.1 covers images and up to 4K at 3 to 10 seconds.
Sources
Related posts
More in Models
- Which Sume video row takes 1:1, 3:4 or 21:9? Aspect ratio matrix
Seedance and MiniMax rows take 21:9; Wan 3.0 skips it; Kling 3 stops at 16:9, 9:16 and 1:1; Omni and Veo are 16:9 or 9:16 only. The full matrix by row.
- Chinese AI video models by API: Kling, Wan, MiniMax limits and prices
Which Chinese video models can you call on Sume? kling-3, wan-3.0, minimax-h3 and minimax-h3-max: clip length, resolution, audio, references and price a second.
- Clef-omni reads video: a cheap check before a second video job
Cloudflare's Clef-omni takes audio and video input at $0.15 per million tokens. Sume does not offer Clef, so this is a pre-check pattern to run outside Sume.
- Eleven v4 on Sume lists 44 language codes, not 90+
ElevenLabs says v4 speaks 90+ languages. Sume's TTS Router catalog lists 44 ISO codes for eleven-v4, and 28 for Multilingual v2. Check yours before you send.
Written by Sume