README.md

metadata

license: apache-2.0
base_model: Tongyi-MAI/Z-Image
base_model_relation: adapter
pipeline_tag: text-to-image
tags:
  - text-to-image
  - lora
  - diffusers
  - z-image
  - zimage
  - stationthis
instance_prompt: stationthis
training_steps: 3000
network_type: lora
library_name: ai-toolkit
widget:
  - text: >-
      stationthis, stationthis, low poly playstation screenshot style, A
      determined young woman with fiery red pigtails tied…
    output:
      url: samples/sample_000.jpg
  - text: >-
      stationthis, stationthis, low poly playstation screenshot style, Under a
      misty, twilight sky, a male adventurer in blue…
    output:
      url: samples/sample_001.jpg
  - text: >-
      stationthis, stationthis, low poly playstation screenshot style, A stoic,
      sunglasses-wearing man in a denim jacket stan…
    output:
      url: samples/sample_002.jpg
  - text: >-
      stationthis, stationthis, low poly playstation screenshot style, A stern,
      determined woman with a black headband and ta…
    output:
      url: samples/sample_003.jpg
  - text: stationthis
    output:
      url: samples/sample_004.jpg
  - text: stationthis
    output:
      url: samples/sample_005.jpg
  - text: stationthis
    output:
      url: samples/sample_006.jpg
  - text: stationthis
    output:
      url: samples/sample_007.jpg

impresstation-zimage

zimage LoRA "impresstation-zimage" — trigger "stationthis".

Trigger word: stationthis

Sample Outputs

stationthis, stationthis, low poly playstation screenshot style, A determined young woman…	stationthis, stationthis, low poly playstation screenshot style, Under a misty, twilight…
:---:	:---:
stationthis, stationthis, low poly playstation screenshot style, A stoic, sunglasses-wear…	stationthis, stationthis, low poly playstation screenshot style, A stern, determined woma…
:---:	:---:
:---:	:---:

Usage (Diffusers)

import torch
from diffusers import DiffusionPipeline

pipe = DiffusionPipeline.from_pretrained(
    "Tongyi-MAI/Z-Image-Turbo", torch_dtype=torch.bfloat16
).to("cuda")
pipe.load_lora_weights("ms2stationthis/impresstation-zimage")
image = pipe("stationthis, a character portrait", guidance_scale=4.0, num_inference_steps=25).images[0]
image.save("out.png")

Recommended Settings

LoRA strength	Guidance	Steps	Resolution
0.8–1.0	4.0	25	1024×1024

Training Details

• Base: Tongyi-MAI/Z-Image
• Steps: 3000 · Network: LoRA rank 32 / alpha 32
• Optimizer: adamw8bit, lr 1e-4 · Scheduler: flowmatch
• Resolution: 512, 768, 1024 (multi-res bucketed) · Precision: bf16 train / fp16 save

Reproduction

This repo includes the full dataset/ (33 image-caption pairs) and the exact config.yaml so the LoRA can be retrained as-is.

About

Trained on StationThis — an AI creative platform powered by $MS2. Train your own LoRAs via @stationthisbot on Telegram.

_{Published via noema.}