
# chisa-fibo

## 模型介绍

本模型依托魔搭社区（ModelScope）AIGC专区[模型训练](https://modelscope.cn/aigc/modelTraining)环境与算力完成训练。

* 模型类型：LoRA
* 基础模型：[briaai/Fibo-2.0-dev](https://modelscope.cn/models/briaai/Fibo-2.0-dev)
* 训练代码：[DiffSynth-Studio](https://github.com/modelscope/DiffSynth-Studio)
* 训练数据量：78
* 总训练步数：10000
* 开源协议：Apache-2.0

## 推理代码

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

```bash
pip install diffsynth
```

开始推理：

```python
import torch
from diffsynth.pipelines.fibo import FiboPipeline, ModelConfig

pipe = FiboPipeline.from_pretrained(
    torch_dtype=torch.bfloat16,
    device="cuda",
    model_configs=[
        ModelConfig(model_id="briaai/Fibo-2.0-dev", origin_file_pattern="transformer/*.safetensors"),
        ModelConfig(model_id="briaai/Fibo-2.0-dev", origin_file_pattern="text_encoder/model.safetensors"),
        ModelConfig(model_id="briaai/Fibo-2.0-dev", origin_file_pattern="vae/diffusion_pytorch_model.safetensors"),
    ],
    processor_config=ModelConfig(model_id="briaai/Fibo-2.0-dev", origin_file_pattern="processor/"),
    font_config=ModelConfig(model_id="briaai/Fibo-2.0-dev", origin_file_pattern="fonts/"),
)
pipe.load_lora(pipe.dit, ModelConfig(model_id="Devilworld/chisa-fibo", origin_file_pattern="chisa-fibo_c1-st10000.safetensors"))

prompt = "a cat"
image = pipe(prompt=prompt, height=1024, width=1024, seed=0, num_inference_steps=50, cfg_scale=5.0)
image.save("image.png")
```


Devilworld/chisa-fibo

chisa-fibo_c1-st3000.safetensors

c1-st3000

c1-st10000

c1-st9000

c1-st8000

c1-st7000

c1-st6000

c1-st5000

c1-st4000

c1-st2000

c1-st1000

Files with SHA256: c0d773f0b0809386db050103dd824e0e807f5954da04e2f02b88a258211397fe