Veo 3.1 API: how to call it, and what Sume lists instead
The Veo 3.1 API is Google's, called through the Gemini API with preview model codes. Sume does not list Veo; here are the catalog models with the same inputs.
The Veo 3.1 API is Google's: you call it through the Gemini API with a model code such as veo-3.1-generate-preview, poll a long-running operation, and download the result. It makes 8-second-class clips (4, 6 or 8 seconds) at 720p, 1080p or 4K with audio always on. Sume's video catalog does not currently list a Veo model. If you need Veo itself, use Google's API; if you need the same kinds of inputs from Sume, the table below maps them.
Google facts come from its Veo page, read 2026-09-29. Sume facts come from the Video generation docs and the catalog code behind GET /v1/videos/models, read the same day.
How do I call the Veo 3.1 API?
Google's REST example sends a POST to models/veo-3.1-generate-preview:predictLongRunning with your key in x-goog-api-key, then polls the returned operation until done is true and downloads the video from the result. The three Veo 3.1 codes on that page are veo-3.1-generate-preview, veo-3.1-fast-generate-preview and veo-3.1-lite-generate-preview, and the page marks all three Preview.
curl -s "https://generativelanguage.googleapis.com/v1beta/models/veo-3.1-generate-preview:predictLongRunning" \
-H "x-goog-api-key: $GEMINI_API_KEY" \
-H "Content-Type: application/json" \
-X POST \
-d '{"instances":[{"prompt":"A red convertible on a coastal road at sunset, engine roaring."}]}'Does Sume have Veo 3.1?
Sume's video catalog does not currently list a Veo model. The listed ids include seedance-2.5, seedance-2-mini, seedance-2, seedance-2-fast, kling-3, wan-3.0, grok-imagine-video-1.5, minimax-h3, minimax-h3-max and gemini-omni-flash-1.1. Ask GET /v1/videos/models for the live list before you build on it, because the catalog changes.
The /v1/videos route follows the OpenRouter video shape, so a client written for it changes only the base URL and key. Model ids on Sume are bare catalog ids with no provider prefix.
Which Sume models take the inputs Veo 3.1 takes?
Veo 3.1 takes text, a first image, a last image, up to three reference images, and audio it generates itself. These catalog models accept the matching inputs on Sume.
| Input | Veo 3.1 (Google) | On Sume |
|---|---|---|
| First and last frame | Yes, image plus lastFrame | frame_images on seedance-2.5, seedance-2, kling-3, wan-3.0, minimax-h3, gemini-omni-flash-1.1 |
| Reference images | Up to 3 | input_references on seedance-2.5, wan-3.0 (up to 10), minimax-h3 (up to 9), gemini-omni-flash-1.1 (up to 10) |
| Native audio | Always on | Always on for minimax-h3 and gemini-omni-flash-1.1; generate_audio on kling-3 and Seedance |
| 4K | Yes, 8 s only | gemini-omni-flash-1.1 lists 4K, 3 to 10 s |
| Longest clip | 8 s | 15 s on most models, 30 s on seedance-2.5 and wan-3.0 |
What do I lose by leaving Veo?
Veo-only features stay Veo-only: Google's page describes extending a Veo-generated video by 7 seconds, up to 20 times, and that works only on videos Veo made. Sume does not list an extend call; chaining clips is the workaround. Match the model to the shot, not the brand, and test one short clip first.
Sources
Related posts
More in Models
- Veo 3.1 extend video, and how to chain clips on Sume
Veo 3.1 extends only videos Veo made, 7 seconds at a time. Sume has no extend call, so you chain clips from a last frame or pick a model with a longer limit.
- Veo 3.1 first and last frame: how the API works
Veo 3.1 takes a first image and a last image and generates the motion between them. Here is Google's request shape and where the same input works on Sume.
- Veo 3.1 Lite, Fast or standard: how to choose
Veo 3.1 has three preview models: standard, Fast and Lite. Lite drops 4K and extension; Fast targets speed. What each does, plus Sume's tiered Seedance.
- Veo 3.1 reference images: three max, and the 8-second rule
Veo 3.1 takes up to three reference images, and a request with them must run 8 seconds. Sume does not list Veo; its reference-capable models take more.
Written by Sume