V4 Styles is live!
Give it one reference image and it holds that style exactly, across everything you generate next.
No training needed, and the reference doesn't have to come from Recraft ↓
fine-tuning — Taking a trained model and training it a bit more on your own examples so it gets better at one specific job.
LoRA — A cheap way to fine-tune a model by training a small add-on layer instead of changing all of the model's weights.
Why it matters
Style consistency across a set of generated assets no longer needs a trained LoRAA cheap way to fine-tune a model by training a small add-on layer instead of changing all of the model's weights.Full definition → or fine-tuningTaking a trained model and training it a bit more on your own examples so it gets better at one specific job.Full definition →. One reference image, including one made elsewhere, pins the look for everything generated afterward.
