Downloads · 30 days
6
22% of all-time downloads
ModelMaya/maya-krea2-v1
maya-krea2-v1 is a text-to-image model from ModelMaya. Use it when you need an image from a text prompt. It is set up for diffusers. The card lists the license as apache-2.0.
Downloads · 30 days
6
22% of all-time downloads
All-time downloads
27
Public
Repo size
971 MB
Likes
0
Public
Click a slice to open those files.
.safetensors576 MB · 59%
From the Hugging Face model README
A DreamBooth-LoRA for Krea 2, trained on Krea 2 RAW and shown on Krea 2 Turbo. The samples below were generated with this LoRA on Turbo (8 steps).
Use the token maya to invoke the concept.

"A high-fashion cinematic portrait of maya wearing a gown made of liquid gold, standing in a futuristic neon-lit Tokyo alleyway during a rainstorm."

"A whimsical oil painting of maya as a forest spirit, lounging on a giant glowing mushroom in a mystical woodland filled with floating spores."

"An epic wide shot of maya in astronaut gear, walking across the iridescent crystalline sands of a distant alien planet under three moons."
import torch
from diffusers import Krea2Pipeline
pipe = Krea2Pipeline.from_pretrained("krea/Krea-2-Turbo", torch_dtype=torch.bfloat16).to("cuda")
pipe.load_lora_weights("ModelMaya/maya-krea2-v1")
image = pipe("A high-fashion cinematic portrait of maya wearing a gown made of liquid gold, standing in a futuristic neon-lit Tokyo alleyway during a rainstorm.", num_inference_steps=8, guidance_scale=0.0).images[0]
image.save("output.png")