Cogview4 6B

🚀 CogView4-6B

CogView4-6B 是一個文本到圖像的模型，可根據輸入的文本生成相應的圖像。該模型在多個基準測試中取得了優異的成績，具有較高的圖像生成質量和準確性。

🤗 模型空間 | 🌐 Github 倉庫 | 📜 CogView3 論文

🚀 快速開始

首先，確保你從源代碼安裝 diffusers 庫。

pip install git+https://github.com/huggingface/diffusers.git
cd diffusers
pip install -e .

然後，運行以下代碼：

from diffusers import CogView4Pipeline

pipe = CogView4Pipeline.from_pretrained("THUDM/CogView4-6B", torch_dtype=torch.bfloat16)

# 開啟以減少 GPU 內存使用
pipe.enable_model_cpu_offload()
pipe.vae.enable_slicing()
pipe.vae.enable_tiling()

prompt = "A vibrant cherry red sports car sits proudly under the gleaming sun, its polished exterior smooth and flawless, casting a mirror-like reflection. The car features a low, aerodynamic body, angular headlights that gaze forward like predatory eyes, and a set of black, high-gloss racing rims that contrast starkly with the red. A subtle hint of chrome embellishes the grille and exhaust, while the tinted windows suggest a luxurious and private interior. The scene conveys a sense of speed and elegance, the car appearing as if it's about to burst into a sprint along a coastal road, with the ocean's azure waves crashing in the background."
image = pipe(
    prompt=prompt,
    guidance_scale=3.5,
    num_images_per_prompt=1,
    num_inference_steps=50,
    width=1024,
    height=1024,
).images[0]

image.save("cogview4.png")

✨ 主要特性

推理要求和模型介紹

• 分辨率：寬度和高度必須在 512px 到 2048px 之間，且能被 32 整除，並確保最大像素數不超過 2^21 像素。
• 精度：支持 BF16 / FP32（不支持 FP16，因為會導致溢出，生成全黑圖像）

使用 BF16 精度和 batchsize=4 進行測試，內存使用情況如下表所示：

模型指標

我們在多個基準測試中進行了測試，並取得了以下分數：

DPG-Bench

GenEval

T2I-CompBench

中文文本準確性評估

📄 許可證

該模型根據 Apache 2.0 許可證 發佈。

📚 引用

🌟 如果你覺得我們的工作有幫助，請考慮引用我們的論文並留下寶貴的 Star。

@article{zheng2024cogview3,
  title={Cogview3: Finer and faster text-to-image generation via relay diffusion},
  author={Zheng, Wendi and Teng, Jiayan and Yang, Zhuoyi and Wang, Weihan and Chen, Jidong and Gu, Xiaotao and Dong, Yuxiao and Ding, Ming and Tang, Jie},
  journal={arXiv preprint arXiv:2403.05121},
  year={2024}
}