Black Forest Labs · Video model
FLUX 3 Video in Twistavo AI Studio
FLUX 3 Video is the video model of the FLUX family from Black Forest Labs. In Twistavo AI Studio it makes clips of 5 to 20 seconds from a prompt, a start frame, an end frame, reference images or source videos. Resolutions run through 720p, 1080p, 2k and 4k, and the frame picker covers wide, landscape, square and vertical shapes. Sound can be generated with the picture or switched off. Every run is quoted in credits before you confirm it, and the finished clip is archived privately in your library. Editing an existing clip is handled by a separate tool and is not part of this generator.
Try FLUX 3 VideoFacts checked
Credits
| Variant | Credits |
|---|---|
| FLUX 3 Video | 258 credits for a 5 s clip at 720p |
Shorter clips cost fewer credits. AI model prices lists each model from its shortest clip, so its numbers can be lower than the default-length quotes here.
Credits come from monthly plans and one-time extra credits for subscribers: see Pricing.
Specifications
| Variant | Type | Length | Resolutions | Frame shapes | Inputs | Sound |
|---|---|---|---|---|---|---|
| FLUX 3 Video | Video | 5–20 s | 720p, 1080p, 2k, 4k | 21:9, 2:1, 16:9, 4:3, 1:1, 3:4, 9:16 | end frame (up to 1), reference images (up to 8), start frame (up to 1) and source video (up to 3) | Optional |
Sources
Where it helps
- Start and end frames give control over where the shot begins and where it lands.
- Reference images keep a product, outfit or character consistent across clips.
- Sound is generated with the picture by default, and the vendor lists multilingual dialogue.
- Resolutions reach UHD for deliverables that need a large file.
Limits to know
- Large resolutions and long clips raise the quote; draft small first.
- Sound is on by default; switch it off before you confirm when you plan to score the clip yourself.
- Editing an existing clip is a separate tool and is not part of this generator.
Prompt tips
- Say what each reference is for, for example the jacket from the first image and the street from the second.
- With a first and a last frame, describe the motion between them, not the pictures themselves.
- Put spoken lines in quotes and name the language when the sound option is on.
- Test framing at the smallest resolution, then rerun the keeper at the size you deliver.
Questions
Which inputs can I add?
end frame, reference images, start frame and source video. The composer shows each slot and its limit before you confirm the quote.
How long can a clip be?
Between 5 and 20 seconds. Longer clips cost more credits; the quote updates as you move the duration control.
Can it generate sound?
Yes, the sound option is on by default. Turn it off when you plan to add music or a voice-over later.
Where do finished clips go?
Into your private library, next to every other AI Studio result, where you can download them or reuse a frame as a new start image.
How do I keep a character consistent across clips?
Add the same reference images to every run and name each one by role in the prompt. Keep the style words identical between prompts, and use a frame from the previous clip as the next start frame when shots must join.
Can I combine several references in one run?
Yes. One run accepts several reference images and source videos, and the composer shows the limit of each slot. A few clearly different references are easier to describe in the prompt than many similar ones.
How does it work with the FLUX image models?
They come from the same vendor and have their own pages. Make the start frame, end frame or reference pictures with an image model first, then bring them into the video composer as inputs.
FLUX 3 Video is a trademark of Black Forest Labs. Twistavo is an independent product, not affiliated with or endorsed by Black Forest Labs; it offers the model through its AI provider. Facts on this page come from the AI Studio catalogue and may change with the provider.