This is my first attempt at making a lora. That being said, i am very surprised with the results. It came out better than i expected. it can even take your photorealistic prompts you may have generated already and turn it into anime/semi-realistic variants. Its recommended to use the trigger words: high_budget_anime_movie_style

Other words that may help for a different style is (used all together): key_visual, ultra_detailed_background, 4k, 8k, cinematic_lighting, painterly_environment, soft_cel_shading,

suggested sampler: euler, er_sde, euler_cfg_pp and euler_ancestral_cfg_pp (adds a lot more color)

I'm pretty sure this lora could be used with flux klein base As well. The images i generated were used with klein distilled model.

This LoRA fine-tunes Flux Klein 9B toward a polished anime aesthetic, emphasizing consistent anatomy, crisp outlines, and strong facial expression control. It improves stylistic cohesion while maintaining prompt accuracy. The recommended strength is 1.00.

I appreciate anyone trying out my first lora! feel free to give me feedback :)