Downloads · 30 days
30
37% of all-time downloads
strkyyy/blueprint-technical
blueprint-technical is a text-to-image model from strkyyy. Use it when you need an image from a text prompt. It is set up for diffusers. The card lists the license as openrail++.
This LoRA model transforms any input into a classic engineering blueprint style, featuring white technical line drawings on a deep cyan-blue background with precise orthographic projections and detailed dimension line…
Downloads · 30 days
30
37% of all-time downloads
All-time downloads
81
Public
Repo size
106 MB
Likes
0
Public
Click a slice to open those files.
.safetensors93.1 MB · 88%
From the Hugging Face model README
This LoRA model transforms any input into a classic engineering blueprint style, featuring white technical line drawings on a deep cyan-blue background with precise orthographic projections and detailed dimension lines and labels.
Trigger the effect by including the word blu3prnt in your prompt. For best results, set the strength between 0.6 and 1.0. Use the EMA sampler for a smoother output, adjusting steps as needed to achieve clarity and detail.
Apply blu3prnt to inputs that require technical precision and clarity. Avoid using overly detailed prompts which may overwhelm the model's ability to render clean, simple lines.
blu3prnt --detailed portrait of an engineer in a dimly lit workshop
blu3prnt --modern cityscape with towering skyscrapers at dusk
blu3prnt --close-up of a vintage mechanical clock with intricate gears
blu3prnt --a robot arm performing precise assembly in a cleanroom
blu3prnt --calm and serene evening scene with rolling hills and trees
blu3prnt --abstract representation of digital circuits on grid paper
Activate with blu3prnt. Recommended strength: 0.9.

diffusersfrom diffusers import AutoPipelineForText2Image
import torch
pipe = AutoPipelineForText2Image.from_pretrained("stabilityai/stable-diffusion-xl-base-1.0", torch_dtype=torch.float16).to("cuda")
pipe.load_lora_weights("strkyyy/blueprint-technical")
image = pipe("blu3prnt, your prompt", num_inference_steps=30).images[0]