Brazilian model and influencer.

1.5 version: For best resemblance use together with the pair embedding (LINK) and adjust weight of the LoRA down if needed.

Flux version trained with roughly similar dataset (her Insta photos are grainy and that's replicated in the model output as well). Just testing some new settings, let me know if it doesn't work under some circumstances. Recommended weight is 1.2 - regardless of how I train, Flux models always need some extra weight for consistent performance shrugs