modele/Shuttle 3.1 Aesthetic - v1.0

Shuttle 3.1 Aesthetic - v1.0

9/28/2025

9:19:29 AM

Powiązane słowa kluczowe i tagi

bfloat16,checkpoint,comfyui,diffusers,flux.1 s,fp8,image generation,shuttle 3.1 aesthetic,shuttleai,style,text-to-image ai model,v1.0

A lone astronaut in a white spacesuit stands atop a skyscraper overlooking the vibrant nighttime skyline of New York City, with illuminated buildings and streets below under a moody sky.

A spiraling tower of ornate translucent stone rises from a snow-covered mountain in a frozen wasteland, with pilgrims in dark ceremonial robes ascending icy steps beneath a glowing moon and ethereal northern lights.

A surreal dusk scene showing a monumental bridge overrun by thorned brambles and carnivorous plants, with a lone dark silhouette standing at the bridge's center against a glowing sunset and reflective water below.

A lone figure in a dark coat stands amidst neon-lit bioluminescent grass, facing a large, iridescent eye embedded in a water-cloud vortex in a cyberpunk digital art style.

A red classic sports car parked on a winding mountain road at sunset, showing rear three-quarter, with textured asphalt, glowing taillights, telephone poles on the right, a Japanese roadside sign, and a distant valley city under a colorful sunset sky.

Digital watercolor illustration of a snow-capped stratovolcano towering over a sprawling modern city at sunrise, with warm golden light illuminating upper slopes and cool blue shadows on lower slopes, sharp urban details, and a bright blue sky with scattered clouds.

A minimal white studio with an antique chair below a floating black hole-like sphere with an iridescent violet to amber ring and surrounding debris.

Vintage yellow ice cream van with a giant ice cream cone on the roof driving along a winding coastal highway carved into rocky cliffs at sunrise, with ocean waves crashing below and seagulls flying overhead.

A towering industrial structure with glowing data panels and scaffolding rises above a fog-covered cyberpunk cityscape. A solitary figure stands silhouetted against holographic displays.

A female cyborg wearing a white armor and futuristic helmet with a glowing red tinted visor stands in a mysterious forest surrounded by glowing orange butterflies and morning light.

Zalecane podpowiedzi

A cat holding a sign that says hello world

Zalecane parametry

steps

4 - 6

resolution

1024x1024, 1280x768

Wskazówki

Enable model CPU offload to save VRAM if needed.

Use torch.compile for potential performance boosts on compatible GPUs, but note it can increase loading times.

Use manual seeds for reproducible image generation results.

Sponsorzy twórcy

Join our Discord to get the latest updates, news, and more.

Try out the model through a website at https://designer.shuttleai.com/

You can use Shuttle 3.1 Aesthetic via API through ShuttleAI and check docs at https://docs.shuttleai.com/

# 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.

![image/png](https://huggingface.co/shuttleai/shuttle-3.1-aesthetic/resolve/main/demo.png)

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.