Instagram to LoRA by https://oneshotlora.com
Source: https://www.instagram.com/dollblush

Unfortunately, civit ai doesn't seem to be able to extract the prompt metadata from videos. The Flux & SDXL images have similar prompts, though, because they were all generated by https://www.oneshotlora.com from the same dataset. The trigger is required, but any of the following in your prompt will help:

D011B, 1girl, solo, lips, realistic, long_hair, looking_at_viewer, blonde_hair, portrait, freckles, smile, jewelry, closed_mouth, parted_lips, blue_eyes, teeth, nose, earrings, necklace, close-up, black_eyes, sitting, white_hair, shirt, upper_body, simple_background, choker, animal_ears, indoors, thighhighs, midriff, rabbit_ears, skirt, glasses, bare_shoulders, holding, dress, one_eye_closed, stuffed_toy, brown_eyes, dark_skin, striped, crop_top, selfie, cellphone, dark-skinned_female, hat, bracelet, holding_phone, lying, smartphone, grey_hair, white_shirt, black-framed_eyewear, fake_animal_ears, bed, closed_eyes, pantyhose, purple_background, short_sleeves, pillow, shorts, standing, stuffed_animal, white_thighhighs, braid, cowboy_shot, heart, long_sleeves, mole, on_bed, brown_hair, eyelashes, hairband, pink_hair, 1boy, flower, hood, horns, leotard, ring