FLUX.1-dev-LoRA-live-3D

This is a LoRA (Live 3D) trained on FLUX.1-dev for 3D elements by Hongke on Shakker AI.

Showcases

Trigger words

Trigger is not required. The recommended scale is 0.9 to 1.3 in diffusers.

Inference

import torch
from diffusers import FluxPipeline

pipe = FluxPipeline.from_pretrained("black-forest-labs/FLUX.1-dev", torch_dtype=torch.bfloat16)
pipe.load_lora_weights("Shakker-Labs/FLUX.1-dev-LoRA-live-3D", weight_name="FLUX-dev-lora-live_3D.safetensors")
pipe.fuse_lora(lora_scale=1.1)
pipe.to("cuda")

prompt = "A colorful cartoon monkey sits on a bus as it rolls down the street in Times Square, New York"

image = pipe(prompt, 
             num_inference_steps=24, 
             guidance_scale=3.5,
            ).images[0]
image.save(f"example.png")

Online Inference

You can also download this model at Shakker AI, where we provide an online interface to generate images.

Acknowledgements

This model is trained by our copyrighted users Hongke. We release this model under permissions. The model follows flux-1-dev-non-commercial-license.