Instructions to use Nanite-Labs/nanites-keypunch-z-image-1.0 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use Nanite-Labs/nanites-keypunch-z-image-1.0 with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("Tongyi-MAI/Z-Image", dtype=torch.bfloat16, device_map="cuda") pipe.load_lora_weights("Nanite-Labs/nanites-keypunch-z-image-1.0") prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k" image = pipe(prompt).images[0] - Inference
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- Draw Things
- DiffusionBee
nanites-keypunch-z-image-1.0
LoRA adapter for Z-Image (Tongyi-MAI, Apache-2.0), fine-tuned on 20 public-domain reference stills of mid-century data instrumentation and paperwork (FSA/OWI, Harris & Ewing, National Photo Company, HABS/HAER surveys) in the idiom of 1930s-1960s U.S. government documentary film stills.
Stark black-and-white: punched cards, tabulating machines and card sorters, strip-chart recorders, dial gauges, ledgers, dense tabular reports. Harsh light, deep blacks, film grain.
Trigger token
Use keypunch in your prompt to invoke the idiom.
Example prompt (Turbo, 9 steps, guidance 0)
keypunch, black-and-white 1940s government documentary film still, close-up of punched cards in a card sorter, harsh light, deep blacks, film grain
Gallery (Base + LoRA, 40 steps)
Training
base_model: Tongyi-MAI/Z-Image
steps: 600
lora_rank: 16
lora_alpha: 16
resolution: 1024
lr: 0.0001
See SOURCES.md in the repo for the per-file rights basis.
- Downloads last month
- -
Model tree for Nanite-Labs/nanites-keypunch-z-image-1.0
Base model
Tongyi-MAI/Z-Image





