Robots Atlas>ROBOTS ATLAS
FLUX 3
AI Models›FLUX

FLUX 3

3 · Family: FLUX
FLUX 3 is Black Forest Labs' generative video model from the FLUX family; thanks to large-scale video pre-training it understands the dynamics of the world. It is the base of the FLUX-mimic Video-Action Model.
⏳ Preview⏳ Limited accessVideo generationWorld Model📁 FLUX
🏢Black Forest LabsProducer
Access:APIDeployment:☁ Cloud

Overview

FLUX 3 is a generative video model developed by Black Forest Labs, part of the FLUX model family. Through large-scale video pre-training the model learns the dynamics and behavior of the world, which — as BFL co-founder and CEO Robin Rombach emphasizes — enables not only understanding the world but also acting in it.

FLUX 3 is the base of the FLUX-mimic Video-Action Model, developed together with mimic robotics: the generative video model is augmented with an action decoder, so that knowledge of the world's dynamics is used to predict robot actions and speed up industrial robot learning.

The model fits Black Forest Labs' mission of building "visual intelligence" — models that understand, reason and act in the world — and extends the FLUX family (which includes FLUX.1 and FLUX.2) into video generation.

Classification
Video generationWorld Model
Family: FLUX
Access & deployment
API
Cloud
Weights: Closed
Key parameters
📥 Input: text, image

Technical specification

Modalities
⬇ Input
textimage
⬆ Output
video

Capabilities and applications

Native model capabilities
Text-to-video generation
Generating video sequences from a text prompt (and optionally an image), with coherent scene dynamics.
Category: video
Image-to-video
The model's ability to animate a static input image — extending it in time into a consistent video clip according to a description of motion or action.
Category: video

Technical architecture

Deployment and security

💾 Related software
💾FLUX-mimicSoftware

Sources and related pages