# Shuttle 3.1 AestheticJoin our [Discord](<a target="_blank" rel="ugc" href="https://discord.gg/shuttleai">https://discord.gg/shuttleai</a>) to get the latest updates, news, and more.## Model VariantsThese model variants provide different precision levels and formats optimized for diverse hardware capabilities and use cases- [bfloat16](<a target="_blank" rel="ugc" href="https://huggingface.co/shuttleai/shuttle-3.1-aesthetic/resolve/main/shuttle-3.1-aesthetic.safetensors">https://huggingface.co/shuttleai/shuttle-3.1-aesthetic/resolve/main/shuttle-3.1-aesthetic.safetensors</a>)- [fp8](<a target="_blank" rel="ugc" href="https://huggingface.co/shuttleai/shuttle-3.1-aesthetic/resolve/main/fp8/shuttle-3.1-aesthetic-fp8.safetensors">https://huggingface.co/shuttleai/shuttle-3.1-aesthetic/resolve/main/fp8/shuttle-3.1-aesthetic-fp8.safetensors</a>)- 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](<a target="_blank" rel="ugc" href="https://huggingface.co/shuttleai/shuttle-3.1-aesthetic/resolve/main/demo.png">https://huggingface.co/shuttleai/shuttle-3.1-aesthetic/resolve/main/demo.png</a>)You can try out the model through a website at <a target="_blank" rel="ugc" href="https://designer.shuttleai.com/">https://designer.shuttleai.com/</a>## Using the model via APIYou can use Shuttle 3.1 Aesthetic via API through ShuttleAI- [ShuttleAI](<a target="_blank" rel="ugc" href="https://shuttleai.com/">https://shuttleai.com/</a>)- [ShuttleAI Docs](<a target="_blank" rel="ugc" href="https://docs.shuttleai.com/">https://docs.shuttleai.com/</a>)## Using the model with 🧨 DiffusersInstall or upgrade diffusers```shellpip install -U diffusers```Then you can use <code>DiffusionPipeline</code> to run the model```pythonimport torchfrom 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.# <a target="_blank" rel="ugc" href="http://pipe.transformer.to">pipe.transformer.to</a>(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.<a target="_blank" rel="ugc" href="http://image.save">image.save</a>("shuttle.png")```To learn more check out the [diffusers](<a target="_blank" rel="ugc" href="https://huggingface.co/docs/diffusers/main/en/api/pipelines/flux">https://huggingface.co/docs/diffusers/main/en/api/pipelines/flux</a>) documentation## Using the model with ComfyUITo run local inference with Shuttle 3.1 Aesthetic using [ComfyUI](<a target="_blank" rel="ugc" href="https://github.com/comfyanonymous/ComfyUI">https://github.com/comfyanonymous/ComfyUI</a>), you can use this [safetensors file](<a target="_blank" rel="ugc" href="https://huggingface.co/shuttleai/shuttle-3.1-aesthetic/blob/main/shuttle-3.1-aesthetic.safetensors">https://huggingface.co/shuttleai/shuttle-3.1-aesthetic/blob/main/shuttle-3.1-aesthetic.safetensors</a>).## Training DetailsShuttle 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.

Shuttle 3.1 Aesthetic

shuttle31Aesthetic_v10.safetensors

v1.0

Files with SHA256: 07bc112b206f117a8f5a840191769d5167e7dcb51604258544c6ab0a2d54001d