Quilt 1m Finetuned Sd3.5

🚀 quilt-1m-finetuned-sd3.5

This project is a full rank finetune derived from sd3/unknown-model. It focuses on text - to - image generation, offering high - quality image outputs with specific validation and training settings.

📄 License

The license for this project is other.

✨ Features

• Multiple Modes: Supports text - to - image and image - to - image generation.
• Fine - Tuned Model: Derived from sd3/unknown-model with specific training settings.
• Gallery of Examples: Provides example images for reference.

📦 Installation

No installation steps are provided in the original document.

💻 Usage Examples

Basic Usage

import torch
from diffusers import DiffusionPipeline

model_id = 'Minh-Ha/quilt-1m-finetuned-sd3.5'
pipeline = DiffusionPipeline.from_pretrained(model_id, torch_dtype=torch.bfloat16) # loading directly in bf16

prompt = "A photo-realistic pathology image"
negative_prompt = 'blurry, cropped, ugly'

pipeline.to('cuda' if torch.cuda.is_available() else 'mps' if torch.backends.mps.is_available() else 'cpu') # the pipeline is already in its target precision level
model_output = pipeline(
    prompt=prompt,
    negative_prompt=negative_prompt,
    num_inference_steps=20,
    generator=torch.Generator(device='cuda' if torch.cuda.is_available() else 'mps' if torch.backends.mps.is_available() else 'cpu').manual_seed(42),
    width=1024,
    height=1024,
    guidance_scale=3.0,
).images[0]

model_output.save("output.png", format="PNG")

📚 Documentation

Validation settings

• CFG: 3.0
• CFG Rescale: 0.0
• Steps: 20
• Sampler: FlowMatchEulerDiscreteScheduler
• Seed: 42
• Resolution: 1024x1024
• Skip - layer guidance:

Note: The validation settings are not necessarily the same as the training settings.

You can find some example images in the following gallery: