Zum Inhalt springen
ModelAufgelistetExtern
Herunterladen

ControlNet Zp12 Normal Gen v1

Recommended version of diffusers is 0.20.2 or 0.24.0 with torch 2.

ArtControlNetVersiondiffusion_pytorch_modelBBewertung

Modellquelle

Art
ControlNet
Dateigröße
1,36 GB
Version
vdiffusion_pytorch_model
Aktualisiert
6. Okt. 2026
Downloads
33
Quelle
Hugging Face

Quellenbeschreibung

Recommended version of diffusers is 0.20.2 or 0.24.0 with torch 2.

Usage Example:

import copy
import torch
import requests
from PIL import Image
from diffusers import DiffusionPipeline, EulerAncestralDiscreteScheduler, ControlNetModel

# Load the pipeline
pipeline: DiffusionPipeline = DiffusionPipeline.from_pretrained(
    "sudo-ai/zero123plus-v1.2", custom_pipeline="sudo-ai/zero123plus-pipeline",
    torch_dtype=torch.float16
)
normal_pipeline = copy.copy(pipeline)
normal_pipeline.add_controlnet(ControlNetModel.from_pretrained(
    "sudo-ai/controlnet-zp12-normal-gen-v1", torch_dtype=torch.float16
), conditioning_scale=1.0)
pipeline.to("cuda:0", torch.float16)
normal_pipeline.to("cuda:0", torch.float16)
# Run the pipeline
cond = Image.open(requests.get("https://d.skis.ltd/nrp/sample-data/0_cond.png", stream=True).raw)
genimg = pipeline(
    cond,
    prompt='', guidance_scale=4, num_inference_steps=75, width=640, height=960
).images[0]
normalimg = normal_pipeline(
    cond, depth_image=genimg,
    prompt='', guidance_scale=1, num_inference_steps=50, width=640, height=960
).images[0]
genimg.save("colors.png")
normalimg.save("normals.png")

Quellen

1 Quelle
Hugging FacePrimärVersion eab5dce4c9a66481fc9e08fe38db58e7e96c8860
huggingface.co/sudo-ai/controlnet-zp12-normal-gen-v1/tree/eab5dce4c9a66481fc9e08fe38db58e7e96c8860

Verifiziert 6. Okt.

Modellartefakte

1 Artefakt

diffusion_pytorch_model.safetensors

safetensors · 1,36 GB · SHA-256 40c9b62a5eaf…4272 · Hugging Face

Download file

Quellenauszüge

3 Auszüge

--- license: cc-by-nc-4.0 --- Recommended version of `diffusers` is `0.20.2` or `0.24.0` with `torch` `2`. Usage Example: ```python import copy import torch import requests from PIL import Image from diffusers import DiffusionPipeline, EulerAncestralDiscreteScheduler, ControlNetModel # Load the pipeline pipeline: DiffusionPipeline = DiffusionPipeline.from_pretrained( "sudo-ai/zero123plus-v1.2", custom_pipeline="sudo-ai/zero123plus-pipeline", torch_dtype=torch.float16 ) normal_pipeline = copy.copy(pipeline) normal_pipeline.add_controlnet(ControlNetModel.from_pretrained( "sudo-ai/controlnet-zp12-normal-gen-v1", torch_dtype=torch.float16 ), conditioning_scale=1.0) pipeline.to("cuda:0", torch.float16) normal_pipeline.to("cuda:0", torch.float16) # Run the pipeline cond = Image.open(requests.get("https://d.skis.ltd/nrp/sample-data/0_cond.png", stream=True).raw) genimg = pipeline( cond, prompt='', guidance_scale=4, num_inference_steps=75, width=640, height=960 ).images[0] normalimg = normal_pipeline( cond, depth_image=genimg, prompt='', guidance_scale=1, num_inference_steps=50, width=640, height=960 ).images[0] genimg.save("colors.png") normalimg.save("normals.png") ```

Source context: 33 downloads · 3 likes · Library diffusers · Repo sudo-ai/controlnet-zp12-normal-gen-v1

Source context: 16 downloads · 3 likes · Library diffusers · Repo sudo-ai/controlnet-zp12-normal-gen-v1