Diffusers How to use Playtime-AI/MM-H3-Amazon_Position_BETA with Diffusers:
pip install -U diffusers transformers accelerate
import torch
from diffusers import DiffusionPipeline
# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("MiniMaxAI/MiniMax-H3", dtype=torch.bfloat16, device_map="cuda")
pipe.load_lora_weights("Playtime-AI/MM-H3-Amazon_Position_BETA")
prompt = "A 10-second 16:9 cinematic shot. integrated_multimodal_description: [Shot 1] Live-action, cinematic, a medium shot frames a naked woman and a naked man on a bed in a softly lit bedroom. She fucks him in the amazon position. She kneels between his raised legs, gripping his thighs firmly and holding them up and apart as she drives her hips forward in a steady, controlled rhythm. Her body rocks with each thrust, her breasts moving with the motion while her hands keep his legs elevated. The man lies on his back beneath her, his legs held high, his torso visible as their bodies connect. Warm practical lamp light casts soft highlights across their skin and gentle shadows along the curves of her hips and his raised legs. The camera holds static with shallow depth of field, keeping sharp focus on their bodies and the explicit motion. overall_soundscape: Wet sex sounds, rhythmic skin-on-skin contact, her steady breathing and soft moans, his deeper breaths, quiet bedroom ambience. non_diegetic_music: N/A"
image = pipe(prompt).images[0] 
