Shuttle 3.1 Aesthetic

# Shuttle 3.1 Aesthetic

Join our [Discord](https://discord.gg/shuttleai) to get the latest updates, news, and more.

## Model Variants

These model variants provide different precision levels and formats optimized for diverse hardware capabilities and use cases

- [bfloat16](https://huggingface.co/shuttleai/shuttle-3.1-aesthetic/resolve/main/shuttle-3.1-aesthetic.safetensors)

- [fp8](https://huggingface.co/shuttleai/shuttle-3.1-aesthetic/resolve/main/fp8/shuttle-3.1-aesthetic-fp8.safetensors)

- GGUF (soon)

Shuttle 3.1 Aesthetic is a text-to-image AI model designed to create detailed and aesthetic images from textual prompts in just 4 to 6 steps. It offers enhanced performance in image quality, typography, understanding complex prompts, and resource efficiency.

You can try out the model through a website at https://designer.shuttleai.com/

## Using the model via API

You can use Shuttle 3.1 Aesthetic via API through ShuttleAI

- [ShuttleAI](https://shuttleai.com/)

- [ShuttleAI Docs](https://docs.shuttleai.com/)

## Using the model with 🧨 Diffusers

Install or upgrade diffusers

```shell

pip install -U diffusers

```

Then you can use DiffusionPipeline to run the model

```python

import torch

from diffusers import DiffusionPipeline

# Load the diffusion pipeline from a pretrained model, using bfloat16 for tensor types.

pipe = DiffusionPipeline.from_pretrained(

"shuttleai/shuttle-3.1-aesthetic", torch_dtype=torch.bfloat16

).to("cuda")

# Uncomment the following line to save VRAM by offloading the model to CPU if needed.

# pipe.enable_model_cpu_offload()

# Uncomment the lines below to enable torch.compile for potential performance boosts on compatible GPUs.

# Note that this can increase loading times considerably.

# pipe.transformer.to(memory_format=torch.channels_last)

# pipe.transformer = torch.compile(

# pipe.transformer, mode="max-autotune", fullgraph=True

# )

# Set your prompt for image generation.

prompt = "A cat holding a sign that says hello world"

# Generate the image using the diffusion pipeline.

image = pipe(

prompt,

height=1024,

width=1024,

guidance_scale=3.5,

num_inference_steps=4,

max_sequence_length=256,

# Uncomment the line below to use a manual seed for reproducible results.

# generator=torch.Generator("cpu").manual_seed(0)

).images[0]

# Save the generated image.

image.save("shuttle.png")

```

To learn more check out the [diffusers](https://huggingface.co/docs/diffusers/main/en/api/pipelines/flux) documentation

## Using the model with ComfyUI

To run local inference with Shuttle 3.1 Aesthetic using [ComfyUI](https://github.com/comfyanonymous/ComfyUI), you can use this [safetensors file](https://huggingface.co/shuttleai/shuttle-3.1-aesthetic/blob/main/shuttle-3.1-aesthetic.safetensors).

## Training Details

Shuttle 3.1 Aesthetic uses Shuttle 3 Diffusion as its base. It can produce images similar to Flux Dev in just 4 steps, and it is licensed under Apache 2. The model was partially de-distilled during training. We overcame the limitations of the Schnell-series models by employing a special training method, resulting in improved details and colors.