infery
← All models

Cosmos Predict 2.5 2B

cosmos-predict-2.5-video-to-video

Video generationby NVIDIA

Generate video from text and videos using NVIDIA's 2B Cosmos Post-Trained Model

Example

Details

Accepts
text + image + video

Pricing

Price
25 cr / video

Prices in credits (1 credit = $0.01).

Data schema

Input

FieldTypeDescription
seedRandom seed for reproducible generation.
promptrequiredstringThe text prompt describing the video to generate.
sync_modebooleanIf `True`, the media will be returned as a data URI and the output data won't be available in the request history.
video_urlstringURL of the input video to use as conditioning.
num_framesintegerNumber of frames to generate. Must be between 9 and 93.
video_qualitystringenum: low, medium, high, maximumThe quality of the output video.
guidance_scalenumberClassifier-free guidance scale. Higher values increase prompt adherence.
negative_promptstringA negative prompt to guide generation away from undesired content.
video_output_typestringenum: X264 (.mp4), VP9 (.webm), PRORES4444 (.mov), GIF (.gif)The format of the output video.
num_inference_stepsintegerNumber of denoising steps. More steps yield higher quality but take longer.
image_urlstringURL of the input image to use as first frame.

Output

FieldTypeDescription
seedintegerThe random seed used for generation.
videoThe generated video file.