Black Forest Labs · video
FLUX 3 Video
Black Forest Labs' video model: up to 20 seconds with audio, up to 4K.
Text to videoImage to videoStart and end frameNative audio
- Length
- 5 to 20 s
- Resolution
- 720p, 1080p, 1440p, 4K
- Aspect ratio
- 16:9, 9:16, 1:1, 4:3, 3:4, 21:9, 2:1
- References
- Up to 2 image files
About FLUX 3 Video
FLUX 3 Video is the FLUX 3 video model from Black Forest Labs, priced at $0.43 per second. It runs as text to video, image to video and start and end frame, with audio, and the length dial goes from 5 seconds to 20 seconds. Output ranges over 720p, 1080p, 1440p and 4K, and the shape can be left to the model or set to 16:9, 9:16, 1:1, 4:3, 3:4, 21:9 and 2:1.
A typical generation takes about 3 min. For a variant tuned for prompt adherence with reasoning before rendering, see FLUX 3 Max Video.
Good for
- Long clips with sound, up to 20 seconds, with a resolution ladder up to 4K
- Transitions between a start and an end frame
Getting good results
- Every resolution above 720p is finished by the provider's video upsampler.
- Sound costs nothing extra here: the audio switch does not change the price.
- The rate climbs with resolution: $2.13 at 720p, $3.63 one step up, $5.88 at 4K.
- Start and end frames can be PNG, JPEG or WebP.
FLUX 3 Video questions
What does the longest clip cost?
$8.50 for 20 seconds at 720p; higher resolutions scale the same rate per second up.