Overview
What is FLUX.3?
FLUX.3 is a Black Forest Labs model available in Starpop. Black Forest Labs' frontier video model for text, image, and start/end frame generation with native audio. Black Forest Labs' frontier video model for cinematic clips from text, one image, or defined first and last frames.
Use text for open exploration, one start frame to preserve a composition, or both frames when the opening and ending states must be controlled. Starpop keeps the model, input mode, duration, format, resolution, and credit choices in one workspace so a team can move from a first test to a publishable asset without rebuilding the brief in another tool.
Specifications
Available in Starpop
These values come from Starpop’s live model configuration and can change as providers add or remove capabilities.
- Inputs
- Text, Image
- Durations
- 5s, 6s, 7s, 8s, 9s, 10s, 11s, 12s, 13s, 14s, 15s, 16s, 17s, 18s, 19s, 20s
- Aspect ratios
- auto, 21:9, 2:1, 16:9, 4:3, 1:1, 3:4, 9:16
- Output options
- 720p, 1080p
- Estimated credits
- 1.7–2.9 credits
- Typical generation time
- About 10 minutes
Best uses
When to choose FLUX.3
Use text for open exploration, one start frame to preserve a composition, or both frames when the opening and ending states must be controlled.
- 01Cinematic text-to-video with native audio
- 02Animating a designed opening frame
- 03Controlled transitions between first and last frames
Considerations
Plan the tradeoffs
1080p uses more credits than 720p, especially on longer clips.
A last frame requires a start frame and a duration that leaves enough time for a coherent transition.
FAQs
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 first and last frames. The right choice still depends on the input assets, delivery format, turnaround, and credit budget for the job.
What inputs does FLUX.3 support in Starpop?
FLUX.3 supports text, image input workflows in Starpop. Available controls are shown in the workspace after the model is selected, so unsupported combinations are not presented as options.
How many credits does FLUX.3 use?
The currently configured estimate is 1.7–2.9 credits. The exact amount shown in Starpop follows the selected resolution, format, duration, or quality option, so review the workspace estimate before generating.
Prompt tips
Get a more useful first result
- 1Describe one continuous action and its intended camera behavior.
- 2Include dialogue, ambience, and sound effects when native audio is enabled.
- 3Use first and last frames with compatible subjects and composition.
- 4Choose the delivery aspect ratio before preparing reference frames.
Similar video models
Compare another approach
Reviewed 2026-08-07 against Starpop configuration and Black Forest Labs source material.
Developer source