sentence-transformers How to use MoralHazard/GISTsmall with sentence-transformers:
from sentence_transformers import SentenceTransformer
model = SentenceTransformer("MoralHazard/GISTsmall")
sentences = [
"Girl with long black hair, healthy tan, and captivating brown eyes smiles warmly. She wears a vibrant blue one-piece swimsuit, high-cut to elongate her toned thighs and highlight her perky breasts. A cute hairclip holds back some strands, and an earclip adds elegance. Seated on a plain chair with a red lollipop near her slightly open mouth, she exudes playful seduction. The simple chalkboard background keeps the focus on her engaging, alluring presence. Overall atmosphere is sensual and enticing, with a bright, cheerful vibe.",
"In a quiet library, a figure emerges from the shadows. Her hair, a dark cascade, frames her pale face, and her eyes, a deep shade of mahogany, hold a mysterious allure. She wears a vintage dress, its color a muted blue, and her movements are graceful. As she sits beside you, the scent of old books and secrets fills the air. Her voice, soft and cultured, shares tales of forgotten worlds, drawing you into a realm of intellectual seduction.",
"You find yourself on a tropical beach, the sun warming your skin. A beautiful stranger, with a radiant smile, catches your eye. She sits across from you, her swimsuit hugging her curves, and her hair shimmering in the sunlight. As she leans forward, offering you the lollipop, her scent, a mix of coconut and summer breeze, envelops you. Her eyes, deep pools of chocolate, hold a promise of sweet delights, and you can't help but be drawn into her playful game.",
"You stand in front of the camera, feeling a little nervous but excited. The photographer, a friend of yours, encourages you to strike a playful pose, something that shows off your personality. You bite your lip, a mix of shyness and mischief in your eyes, and lean against the wall, one hand on your hip. The outfit clings to your curves, and you can't help but feel a little sexy, a side of yourself you don't often explore."
]
embeddings = model.encode(sentences)
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [4, 4] 
