---
title: Krea-2-LoRA-retroanime
canonical_url: "https://www.modelscope.cn/models/krea/Krea-2-LoRA-retroanime"
md_url: "https://www.modelscope.cn/models/krea/Krea-2-LoRA-retroanime.md"
repository: krea/Krea-2-LoRA-retroanime
last_updated: 2026-07-01
license: other
pipeline_tag: text-to-image-synthesis
tasks:
  - text-to-image-synthesis
base_model:
  - krea/Krea-2-Turbo
base_model_relation: adapter
parameters: 117.3M
tensor_type:
  - F32
library_name:
  - pytorch
  - lora
  - safetensors
frameworks:
  - pytorch
language:
  - en
supports_inference: txt2img
downloads: 157
stars: 3
tags:
  - lora
---

# Krea-2-LoRA-retroanime

> Krea-2-LoRA-retroanime - krea 在 ModelScope 开源的模型。Krea 2 LoRA — retroanime

krea/Krea-2-LoRA-retroanime 是 ModelScope 魔搭社区上的 117.3M 参数text-to-image-synthesis模型，采用 other 许可，基于 krea/Krea-2-Turbo 构建，并支持在线推理（txt2img）。

- **Repository**: krea/Krea-2-LoRA-retroanime
- **License**: other
- **Tasks**: text-to-image-synthesis
- **Parameters**: 117.3M
- **Base model**: krea/Krea-2-Turbo
- **Online inference**: txt2img
- **Tags**: lora
- **Downloads**: 157
- **Stars**: 3
- **Last updated**: 2026-07-01

Source: https://www.modelscope.cn/models/krea/Krea-2-LoRA-retroanime

---

# Krea 2 LoRA — retroanime

A LoRA for [Krea 2](https://huggingface.co/krea) — retroanime (trigger: `Purple retro anime style`).
- **Trigger word:** `Purple retro anime style`
- **To be used on:** [`krea/Krea-2-Turbo`](https://huggingface.co/krea/Krea-2-Turbo), the few-step distilled checkpoint shown in the previews above.
- **Trained on:** [`krea/Krea-2-Raw`](https://huggingface.co/krea/Krea-2-Raw).
- **Weights:** `retroanime.safetensors`
- **Previews:** rendered on Turbo at 8 steps, guidance 0.0, LoRA weight `1.0`.

## Usage

```python
import torch
from diffusers import Krea2Pipeline

pipe = Krea2Pipeline.from_pretrained("krea/Krea-2-Turbo", torch_dtype=torch.bfloat16).to("cuda")
pipe.transformer.load_lora_adapter("krea/Krea-2-LoRA-retroanime", weight_name="retroanime.safetensors")
pipe.transformer.set_adapters("default", weights=1.0)

prompt = "A deer grazing in the forest, Purple retro anime style"
image = pipe(prompt, num_inference_steps=8, guidance_scale=0.0).images[0]
image.save("retroanime.png")
```

<Gallery />