FLUX.2
The FLUX lineage, now partly Apache
What it is
Black Forest Labs' successor to FLUX.1, covering text-to-image plus single- and multi-reference editing in one architecture. The distilled klein 4B runs on ~8GB VRAM under Apache-2.0, while the 32B dev model sets the open-weights quality bar under a non-commercial license.
Why it's interesting
The most prominent Western open-weights image family — and klein 4B is the first genuinely Apache-licensed FLUX-quality model that runs on mainstream consumer GPUs. The catch: the family mixes licenses per variant.
Use cases
- High-end text-to-image
- Reference-based editing and character consistency
- Local generation with klein
Who it's for
Prosumers, researchers, and (for klein 4B) commercial users
Setup
Moderate. klein 4B: ~8GB VRAM. dev 32B: H100-class, or a 4090 with FP8 quantization
Limitations & cautions
The flagship dev weights are non-commercial; distilled variants trade peak quality for speed.
Editorial takeaway
A family portrait of the open-weights business model: give away the small one, license the big one. Read the label on each box.