You're looking at a specific version of this model. Jump to the model overview.

black-forest-labs /flux-3:ce1bbc2e

Input schema

The fields you can use to run this model with an API. If you don’t give a value for a field its default value will be used.

Field Type Default value Description
prompt
string
Text description of the video to generate. Plain language works — the prompt is interpreted and expanded before generation. Describe the scene, action, camera moves, and any audio you want.
image
string
A single image that opens the clip, shown pixel for pixel as the first frame (image-to-video). Combine with end_image to morph between two images. Leave empty for text-to-video. Must be PNG, JPEG, or WebP.
end_image
string
Optional end image. When set together with image, the model morphs from image to end_image and fills the in-between frames. Requires a whole-number duration. Must be PNG, JPEG, or WebP.
keyframe_images
array
[]
A storyboard of up to 10 images the model connects into one clip, evenly spaced across the duration. Use this instead of image/end_image for more than two keyframes. Must be PNG, JPEG, or WebP.
reference_images
array
[]
1 to 10 images that define who or what appears. The subjects stay recognizable and the model builds a new scene around them; the images themselves never appear on screen. Must be PNG, JPEG, or WebP.
reference_video
string
A video whose subjects and identity are carried over into a brand-new scene and action. Must be an mp4, at most 50MB and 15 seconds.
start_video
string
A video to continue from its final frames. Use this to extend a shot or chain generations into a longer sequence. Must be an mp4, at most 50MB and 15 seconds.
aspect_ratio
None
auto
Aspect ratio of the generated video. 'auto' picks a ratio from your prompt and any inputs.
resolution
None
720p
Output resolution. With a video input, 720p tops out at 15 seconds — use 480p for durations up to 20 seconds.
duration
None
auto
Length of the generated clip in seconds. 'auto' lets the model pick (5, 10, 15, or 20). A two-image morph (image + end_image) needs a whole-number duration.
generate_audio
boolean
True
Generate synchronized audio (ambient sound, speech, effects). Set to false for a silent clip.
seed
integer
Random seed for reproducible generation. Leave empty for a random seed.

Output schema

The shape of the response you’ll get when you run this model with an API.

Schema
{'format': 'uri', 'title': 'Output', 'type': 'string'}