import torch
from diffusers import DiffusionPipeline
# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("CelesteImperia/SDXL-MLX", dtype=torch.bfloat16, device_map="cuda")
prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k"
image = pipe(prompt).images[0]CelesteImperia: SDXL MLX (Apple Silicon Native)
Harness the power of the Apple M-series (M1-M4) Unified Memory architecture. This spoke provides the Vedic Cinematic experience with zero overhead.
🍎 Why MLX?
- Unified Memory: Seamlessly shares system RAM for massive 1024x1024 textures.
- Native Format: Weights are structured in
.safetensorsspecifically for the MLX framework. - Efficiency: Optimized for Metal (Apple GPU) and the Apple Neural Engine.
📦 Components
unet/: High-precision Safetensors.tinyvae/: MLX-structured Encoder/Decoder for rapid latent-to-image conversion.
🛠️ Quick Start
Ideal for use with mlx-examples or the diffusers MLX backend.
- Downloads last month
- -
Hardware compatibility
Log In to add your hardware
Quantized
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support
Model tree for CelesteImperia/SDXL-MLX
Base model
stabilityai/stable-diffusion-xl-base-1.0