Meet FLUX 3
FLUX 3 is the video model from Black Forest Labs, the team behind the FLUX image models. It generates from a prompt alone, animates a start frame while preserving its composition, or moves between a first and a last frame you designed, over anywhere from 5 to 20 seconds.
There are two tiers. Draft is a much cheaper 720p preview for testing prompts; Standard adds 1080p for the final render. Native audio can be switched off when you plan to score the clip yourself.
Available in Advibly
Settings and estimated costs from Advibly's live model configuration.
- Inputs
- Text, Start frame, End frame
- Durations
- 5s to 20s
- Aspect ratios
- 9:16, 16:9, 4:3, 3:4, 1:1, 21:9, 2:1
- Output options
- 720p, 1080p
- Quality tiers
- Draft, Standard
- Audio
- Generated with the video, can be switched off
- Estimated cost
- $0.09 to $0.44 / secondOne credit costs $1. The range spans the cheapest and the most expensive settings this model offers; the composer shows the exact amount before you generate.0.09 to 0.44 credits / second
When to use it
Use text for open exploration, one start frame to preserve a composition, or both frames when the opening and ending states must be controlled.
- Cinematic text-to-video with native audio
- Animating a designed opening frame
- Controlled transitions between a first and a last frame
Things to keep in mind
1080p is only on the Standard tier and costs more than 720p, especially on longer clips.
A last frame needs a start frame and a duration long enough for a coherent transition.
Get more from your prompt
- 1.Describe one continuous action and the camera's behaviour during it.
- 2.Include dialogue, ambience and sound effects when audio is on.
- 3.Use first and last frames with matching subjects and composition.
- 4.Test on Draft, then re-run the same prompt on Standard.
Common questions
What is FLUX 3 best used for?
FLUX 3 is particularly useful for cinematic text-to-video with native audio, animating a designed opening frame, controlled transitions between a first and a last frame. The right choice still depends on your input assets, delivery format, turnaround and credit budget.
What inputs does FLUX 3 support in Advibly?
FLUX 3 works from text, start frame and end frame. Advibly only shows the controls a model actually supports once it is selected, so you are never offered a combination it cannot render.
How many credits does FLUX 3 use?
The current estimate is 0.09 to 0.44 credits / second ($0.09 to $0.44 / second, at $1 per credit). The exact amount follows the options you pick, and Advibly shows it on the generate button before anything is charged.
Reviewed September 29, 2026 against Advibly's model configuration and Black Forest Labs's published material.
Developer source
