import torch
from diffusers import DiffusionPipeline
# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("Balux/budapest-suburban-train", dtype=torch.bfloat16, device_map="cuda")
prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k"
image = pipe(prompt).images[0]Budapest Suburban Train Rebirth
The idea here was to take the (in-)famous suburban train of Budapest, (the LEW MXA or as we call it HÉV that is in operation since 1971) and inject its characteristics into modern transportation vehicles.
Sample pictures of this concept:
.png)
.png)
The instance of the LEW MXA can be invoked by the token "bphev" in the prompt.
The 1st sample image was generated with the followings:
Prompt: bphev that looks like a (scifi) vehicle , sharp, photo realistic, cinematic light, focus, photorealism, beautiful reflections
Negative prompt: blur, blurred, out of frame, low resolution, low quality, cropped, (noise)
Steps: 20, Sampler: Euler a, CFG scale: 7, Seed: 1456692169, Size: 512x512, Model hash: 5f06aa6f
All the rest was created following the above recipe with mild alternations + img2img upscaling.
Test the concept on Google Colab: fast-Colab-A1111
- Downloads last month
- 28