FLUX is Black Forest Labs' family of frontier image (and now multi-modal) generation models, available in open and commercial versions and via API.

A single multi-modal model for image, video, audio, and action prediction, jointly trained in one architecture, with FLUX 3 Video in early access.

Black Forest Labs News

FLUX 3 Image Klein

October 15, 2026

A 5.6B parameter open weights model, leading our blind evaluations for text-to-image, multi-reference generation and image editing.

Introducing FLUX 3 Action

September 23, 2026

An open weights 7B World Action Model that achieves first place on the RoboLab benchmark.

FLUX Video Upscale: 2K and 4K

August 20, 2026

FLUX Upscale takes any video and regenerates it at a higher resolution, up to native 4K, now available as its own FLUX Tool and endpoint.

FLUX Upscale: 2K and 4K for Video

August 20, 2026

FLUX Upscale takes any video and regenerates it at a higher resolution, up to native 4K, now available as its own FLUX Tool and endpoint.

FLUX 3 by Black Forest Labs just landed on Picsart, and it covers a lot of creative ground. Generate up to 20 seconds of video with native audio from a text prompt, an image, a reference clip, or defined keyframes, and carry a character from one scene into a completely new one.

View on X

Introducing FLUX 3

July 23, 2026

Black Forest Labs introduced FLUX 3: one multi-modal model for image, video, audio, and action prediction, with creations truer to life across every style. Jointly trained in a single unified architecture, it can even be extended to predict actions for robotics. FLUX 3 Video is now in early access.