---
title: cococo
canonical_url: "https://www.modelscope.cn/models/wosoxo/cococo"
md_url: "https://www.modelscope.cn/models/wosoxo/cococo.md"
repository: wosoxo/cococo
chinese_name: cocoo
last_updated: 2026-06-22
license: "Apache License 2.0"
pipeline_tag: any-to-any
tasks:
  - any-to-any
parameters: 707.8M
tensor_type:
  - BF16
library_name:
  - pytorch
  - safetensors
frameworks:
  - Pytorch
downloads: 72
stars: 2
---

# cococo

> cococo - wosoxo 在 ModelScope 开源的模型。模型 ID：cococo展示名称：cococo框架：PyTorch任务类型：多模态 / 统一多模态开源协议：Apache License 2.0公开类型：公开模型二、模型介绍本模型基于 PyTorch 搭建，依托 COCO 数据集完成训练，是一款通用统一多模态模型，支持图像、文本多模态特征融合，可用于目标检测、图文匹配、视觉内容理解等下游任务，原生兼容 COCO 标准数据集格式，自带完整训练、验证、评测全流程能力。

wosoxo/cococo 是 ModelScope 魔搭社区上的 707.8M 参数any-to-any模型，采用 Apache License 2.0 许可。

- **Repository**: wosoxo/cococo
- **License**: Apache License 2.0
- **Tasks**: any-to-any
- **Parameters**: 707.8M
- **Downloads**: 72
- **Stars**: 2
- **Last updated**: 2026-06-22

Source: https://www.modelscope.cn/models/wosoxo/cococo

---

# cococo

## 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：[Qwen/Qwen-Image-2512](https://modelscope.cn/models/Qwen/Qwen-Image-2512)
* 训练代码：[DiffSynth-Studio](https://github.com/modelscope/DiffSynth-Studio)
* 训练数据量：10
* 总训练步数：6000
* 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
from diffsynth.pipelines.qwen_image import QwenImagePipeline, ModelConfig
import torch

pipe = QwenImagePipeline.from_pretrained(
    torch_dtype=torch.bfloat16,
    device="cuda",
    model_configs=[
        ModelConfig(model_id="Qwen/Qwen-Image-2512", origin_file_pattern="transformer/diffusion_pytorch_model*.safetensors"),
        ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="text_encoder/model*.safetensors"),
        ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="vae/diffusion_pytorch_model.safetensors"),
    ],
    tokenizer_config=ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="tokenizer/"),
)
pipe.load_lora(pipe.dit, ModelConfig(model_id="wosoxo/cococo", origin_file_pattern="cococo_c1-st6000.safetensors"))
prompt = "a cat"
image = pipe(prompt)
image.save("image.jpg")
```
