All models
Black Forest LabsbestText → VideoImage → Video

FLUX 3 Video

black-forest-labs/flux-3-video

Estimated

$0.50 / 5s @ 720p

Generates video from a text prompt (t2v), animates one or more reference images (i2v) or continues an existing clip (v2v). Synchronized audio, up to FHD, 5–20s.

Audio HD/FHD 5–20s Up to 10 images Draft mode

Video URLs are signed and expire after ~2 hours.

Input schema

The exact fields accepted by this model. Values are validated before a generation starts.

modestringrequireddefault: "t2v"

Generation mode.

t2v · Text → Videoi2v · Image → Videov2v · Video → Video
promptstringrequired

Free-form prompt describing the video.

keyframesimage

For i2v: one or more reference images (HTTPS URL or base64). Optional timed pairs like [0, image], [3.5, image].

start_videostring

For v2v: the video clip to continue (URL or base64 mp4, ≤50MB, ≤15s).

aspect_ratiostringdefault: "auto"

'auto' picks the best ratio from the prompt or references.

auto · Auto21:92:116:94:31:13:49:16
durationintegerdefault: 5

Clip length in seconds, or 'auto'.

Range: 520

resolutionstringdefault: "hd"

Output resolution.

hd · HDfhd · FHD
generate_audiobooleandefault: true

Synchronized audio.

draftbooleandefault: false

Fast, low-cost preview (HD only).

safety_toleranceintegerdefault: 2

Harm moderation tolerance (0–4).

Range: 04

Default values

{
  "mode": "t2v",
  "aspect_ratio": "auto",
  "duration": 5,
  "resolution": "hd",
  "generate_audio": true,
  "draft": false,
  "safety_tolerance": 2
}