Flux 3 video generator
- Type
- Video model
- Made by
- Black Forest Labs
- Input
- Text, plus a starting image, or a first and last frame
- Output
- MP4 video, with or without sound
- Options
- 7 aspect ratios · Duration 5–20 s · Resolution: 720p, 1080p · Sound · First & last frame
- Price
- From 2,125 credits
- Plan
- Paid plans
Flux 3 is Black Forest Labs' video model. It renders clips of up to 20 seconds in a single pass, with sound built in, from a text prompt, a starting image, or a first and last frame.
What you can do with it
- Text to video. Describe a shot and get a clip of 5 to 20 seconds.
- Start from an image. Attach a picture and the first image is used as the opening frame.
- First and last frame. Switch on First & last frame, add the first frame (required) and an optional last frame, and the clip moves from one to the other.
- Sound. Leave Sound on to get audio with the clip, or switch it off.
How to use it
- Write the shot in the composer.
- Attach a starting image, or switch on First & last frame and add your frames.
- Set aspect ratio, duration, resolution and sound with the chips.
- Check the price in the composer, then generate.
Writing prompts for Flux 3
A 20-second clip gives an action room to play out, so it helps to describe how the shot develops.
- Describe a beginning and an end. "The kettle starts to steam, then the camera pulls back to reveal the whole kitchen."
- Keep it to one location. One continuous shot in one place holds together better than a jump between scenes.
- Give the camera one clear move. A slow orbit, a dolly in, a crane up over the rooftops.
- Write the soundtrack. Footsteps on gravel, wind in tall grass, a low synth drone, no dialogue.
- From a starting image, describe the motion and sound, not what's already in the picture.
Example prompts:
- A paper boat drifts down a rain-filled gutter, bobbing past leaves and bottle caps. Low tracking shot at water level. Rain patter, distant thunder.
- An old lighthouse at dusk, its lamp turning on as the camera slowly rises from the rocks to the top of the tower. Waves crashing, wind, a single gull.
Choosing settings
- Aspect ratio: 16:9 (default), 9:16, 4:3, 3:4, 1:1, 21:9 or 2:1.
- Duration: 5 to 20 seconds, default 8.
- Resolution: 720p (default) or 1080p.
- Sound: on by default.
- First & last frame: off by default.
Pricing is per second of video: 425 credits at 720p and 725 credits at 1080p. Multiply by the duration: a 5-second 720p clip is 2,125 credits, an 8-second 720p clip is 3,400, and a full 20-second clip at 1080p is 14,500. Try your idea at a short duration and 720p first, then generate the longer or sharper version once the shot works. The composer shows the exact price before you generate.
For short, low-cost drafts, LTX 2.3 also reaches 20 seconds. For clips up to 30 seconds, see Wan 3.0.