---
title: CFW_ZIB_test1
canonical_url: "https://www.modelscope.ai/models/Timothy_59/CFW_ZIB_test1"
md_url: "https://www.modelscope.ai/models/Timothy_59/CFW_ZIB_test1.md"
repository: Timothy_59/CFW_ZIB_test1
last_updated: 2026-06-01
pipeline_tag: text-to-image-synthesis
tasks:
  - text-to-image-synthesis
parameters: 635.0M
tensor_type:
  - BF16
library_name:
  - pytorch
  - safetensors
frameworks:
  - pytorch
supports_inference: txt2img
downloads: 348
stars: 0
base_model: Tongyi-MAI/Z-Image
license: "Apache License 2.0"
tags:
  - LoRA
trigger_words: "cl3ric, f1ght3r, w1tch"
vision_foundation: Z_IMAGE_TURBO
---

# CFW_ZIB_test1

> CFW_ZIB_test1 - An open-source model by Timothy_59 on ModelScope. Model Introduction This was intended to test making a 3 character lora based on phtorealistic AI-generated TTRPG character art. There is a female cleric with red hair (trigger: cl3r1c), a male…

Timothy_59/CFW_ZIB_test1 is a 635.0M-parameter text-to-image-synthesis model on ModelScope. and supports online inference (txt2img).

- **Repository**: Timothy_59/CFW_ZIB_test1
- **Tasks**: text-to-image-synthesis
- **Parameters**: 635.0M
- **Online inference**: txt2img
- **Downloads**: 348
- **Stars**: 0
- **Last updated**: 2026-06-01

Source: https://www.modelscope.ai/models/Timothy_59/CFW_ZIB_test1

---

# 

## Model Introduction
This was intended to test making a 3 character lora based on phtorealistic AI-generated TTRPG character art. There is a female cleric with red hair (trigger: cl3r1c), a male fighter (trigger: f1ght3r) and a female witch with white hair (trigger: w1tch). The results could be better. The women bleed together when in the same scene. Perhaps more training is needed, or perhaps Z Image models will always have too much character bleed even when trained in one lora.

This model is trained using the [Civision/model training](https://modelscope.ai/civision/modelTraining) environment and computing resources provided by ModelScope.

* Model Type: LoRA
* Base Model: [Tongyi-MAI/Z-Image](https://modelscope.ai/models/Tongyi-MAI/Z-Image)
* Training Code: [DiffSynth-Studio](https://github.com/modelscope/DiffSynth-Studio)
* Data amount：75
* Total training steps：8000
* License: Apache-2.0

## Inference Code

Install [DiffSynth-Studio](https://github.com/modelscope/DiffSynth-Studio):

```bash
pip install diffsynth
```

Inference:

```python
from diffsynth.pipelines.z_image import ZImagePipeline, ModelConfig
import torch

pipe = ZImagePipeline.from_pretrained(
    torch_dtype=torch.bfloat16,
    device="cuda",
    model_configs=[
        ModelConfig(model_id="Tongyi-MAI/Z-Image", origin_file_pattern="transformer/*.safetensors"),
        ModelConfig(model_id="Tongyi-MAI/Z-Image", origin_file_pattern="text_encoder/*.safetensors"),
        ModelConfig(model_id="Tongyi-MAI/Z-Image", origin_file_pattern="vae/diffusion_pytorch_model.safetensors"),
    ],
    tokenizer_config=ModelConfig(model_id="Tongyi-MAI/Z-Image", origin_file_pattern="tokenizer/"),
)
pipe.load_lora(pipe.dit, ModelConfig(model_id="Timothy_59/CFW_ZIB_test1", origin_file_pattern="CFW_ZIB_test1_c1-st8000.safetensors"))
prompt = "a cat"
image = pipe(prompt=prompt, num_inference_steps=50, cfg_scale=4)
image.save("image.jpg")
```