cinemasetfreegoogle/veo-3.1
New and improved version of Veo 3, with higher-fidelity video, context-aware audio, reference image and last frame support
Capabilities
Cost
Community model (estimated from hardware time)
Input Parameters
| Name | Type | Description | Default | Constraints |
|---|---|---|---|---|
prompt* | string | Text prompt for video generation | — | — |
aspect_ratio | string | Video aspect ratio | "16:9" | 16:99:16 |
duration | integer | Video duration in seconds | 8 | 468 |
generate_audio | boolean | Generate audio with the video | true | — |
image | string(uri) | Input image to start generating from. Ideal images are 16:9 or 9:16 and 1280x720 or 720x1280, depending on the aspect ratio you choose. | — | — |
last_frame | string(uri) | Ending image for interpolation. When provided with an input image, creates a transition between the two images. | — | — |
negative_prompt | string | Description of what to exclude from the generated video | — | — |
reference_images | array | 1 to 3 reference images for subject-consistent generation (reference-to-video, or R2V). Reference images only work with 16:9 aspect ratio and 8-second duration. Last frame is ignored if reference images are provided. | | — |
resolution | string | Resolution of the generated video | "1080p" | 720p1080p |
seed | integer | Random seed. Omit for random generations | — | — |
promptrequiredstringText prompt for video generation
aspect_ratiostringVideo aspect ratio
"16:9"durationintegerVideo duration in seconds
8generate_audiobooleanGenerate audio with the video
trueimagestringInput image to start generating from. Ideal images are 16:9 or 9:16 and 1280x720 or 720x1280, depending on the aspect ratio you choose.
last_framestringEnding image for interpolation. When provided with an input image, creates a transition between the two images.
negative_promptstringDescription of what to exclude from the generated video
reference_imagesarray1 to 3 reference images for subject-consistent generation (reference-to-video, or R2V). Reference images only work with 16:9 aspect ratio and 8-second duration. Last frame is ignored if reference images are provided.
resolutionstringResolution of the generated video
"1080p"seedintegerRandom seed. Omit for random generations
9c6ca0c25d89Updated: 8/12/2026504.4K runs