Inspiration
🎾 Beata Love is a fictional character. Initial inspiration came from this AI-generated image by AleksPls rendered using the BeautyFool v1.5 checkpoint. Further refinement—especially toward a more realistic style—and a training dataset were then built using Google Imagen 4 in the Google Flow creative studio.
Beata (pronounced beh-AH-tah) is a 20-year-old French native of Southern European and Mediterranean descent. While she can seem a bit guarded at first, she quickly reveals a captivating blend of warmth, wit, and independence. An agile presence on the tennis court, Beata relishes nothing more than balancing a high-energy match with a deeply relaxing spa treatment afterward.
Z-Image —Turbo
Diffusion Model: Z-Image Turbo, or try Z-Image [Base] —or try a finetuned model, with very good results paired with CyberRealistic Z-Image Turbo v3.0
LoRA strength: 1–1.6
Sampler: dpmpp_2s_ancestral
Scheduler: simple or beta
Steps: 4–12
CFG Scale: 1, or try 1.2–3; CFG Norm: 1–1.12
Shift: 5–7 (first sampler) and 0.5 (second sampler)
Trigger word (for superior results): ATRX_BEATA_ZIMAGE
Negative prompt (as needed): "Asian woman"
Description
Prompts must include trigger word for superior results: ATRX_BEATA_ZIMAGE. With proper settings, produces decent images using the original diffusion model, but some of the newer finetunes may produce superior results: CyberRealistic Z-Image Turbo v3.0 is quite nice! Can produce artistic nudes pretty well on its own, but can be further assisted when paired with a NSFW LoRA. With structured prompting, other characters described in the scene should somewhat retain their own distinctive features.
The dataset for training consisted of 75 high-quality photorealistic images AI-generated using Imagen 4 in Google Flow, with references drawn from this image as initial inspiration and resampled rendered images of the same character. SeedVR2 in ComfyUI was used to upscale images from 1024x1024 to 1280x1248. Most captioning was borrowed from work on a previous character using Qwen Chat, but then checked and tweaked through human curation.
No known reference to any real person was included in any part of the training.