智谱CogVideoX视频开源大模子
一、资料地址GitHub - THUDM/CogVideo: Text-to-video generation: CogVideoX (2024) and CogVideo (ICLR 2023)
CogVideo/README_zh.md at main · THUDM/CogVideo · GitHub
二、CogVideo部署与实现
方式1-基于源码部署
步调一：下载源码
cd /workspace/
git clone https://github.com/THUDM/CogVideo.git 步调二：下载依靠库
cd /workspace/CogVideo/
pip install -r requirements.txt
cd sat
pip install -r requirements.txt
pip install omegaconf 步调三：下载模子库
mkdir THUDM
cd THUDM
git lfs install
git clone https://www.modelscope.cn/ZhipuAI/CogVideoX-5b-I2V.git 步调四：测试
cd ..
cd inference
python cli_demo.py 方式2-基于Docker部署
步调一：安装docker
apt install podman-docker
apt install docker.io 步调二：启动docker
systemctl start docker
systemctl enable docker 步调三：下载
docker run -itd --name=cogvideo -p 7878:7878 --gpus=all registry.cn-hangzhou.aliyuncs.com/guoshiyin/cogvideo:v3
方式3-基于modelscope调用
魔搭社区
https://i-blog.csdnimg.cn/direct/9890af35ec83406d8a44323c3554c1e6.png
步调一：下载依靠库
pip install modelscope
pip install torch
pip install accelerate
pip install sentencepiece
pip install --upgrade opencv-python transformers
pip install git+https://github.com/huggingface/diffusers.git@878f609aa5ce4a78fea0f048726889debde1d7e8#egg=diffusers # Still in PR 步调二：编写代码
mkdir /workspace/
touch cli.py
vi cli.py 在cli.py中添加代码如下：
# To get started, PytorchAO needs to be installed from the GitHub source and PyTorch Nightly.
# Source and nightly installation is only required until the next release.
import torch
from diffusers import AutoencoderKLCogVideoX, CogVideoXTransformer3DModel, CogVideoXImageToVideoPipeline
from diffusers.utils import export_to_video, load_image
from transformers import T5EncoderModel
from torchao.quantization import quantize_, int8_weight_only
quantization = int8_weight_only
text_encoder = T5EncoderModel.from_pretrained("THUDM/CogVideoX-5b-I2V", subfolder="text_encoder", torch_dtype=torch.bfloat16)
quantize_(text_encoder, quantization())
transformer = CogVideoXTransformer3DModel.from_pretrained("THUDM/CogVideoX-5b-I2V",subfolder="transformer", torch_dtype=torch.bfloat16)
quantize_(transformer, quantization())
vae = AutoencoderKLCogVideoX.from_pretrained("THUDM/CogVideoX-5b-I2V", subfolder="vae", torch_dtype=torch.bfloat16)
quantize_(vae, quantization())
# Create pipeline and run inference
pipe = CogVideoXImageToVideoPipeline.from_pretrained(
"THUDM/CogVideoX-5b-I2V",
text_encoder=text_encoder,
transformer=transformer,
vae=vae,
torch_dtype=torch.bfloat16,
)
pipe.enable_model_cpu_offload()
pipe.vae.enable_tiling()
pipe.vae.enable_slicing()
prompt = "A little girl is riding a bicycle at high speed. Focused, detailed, realistic."
image = load_image(image="input.jpg")
video = pipe(
prompt=prompt,
image=image,
num_videos_per_prompt=1,
num_inference_steps=50,
num_frames=49,
guidance_scale=6,
generator=torch.Generator(device="cuda").manual_seed(42),
).frames
export_to_video(video, "output.mp4", fps=8) 步调三：调试
cd /workspace/
python cli.py https://i-blog.csdnimg.cn/direct/c73f21c4172d42f19ce11b6b362141d9.png
三、视频创作平台
1、清影-智谱
体验地址：智谱清言
智谱AI开放平台
https://i-blog.csdnimg.cn/direct/3fb9858b9cab4ef6a6051357d860c9ea.png
提示词
一只熊猫，穿着一件赤色的小夹克，戴着一顶小帽子，坐在宁静的竹林里的木凳上。熊猫毛茸茸的爪子拨弄着一把微型原声吉他，发出柔和的旋律。附近，其他几只熊猫聚集在一起，好奇地看着，有些还有节奏地鼓掌。阳光透过高大的竹子，在现场投下柔和的光芒。熊猫的脸很有心情，在玩耍时表现出专注和快乐。配景包罗一条小溪和生机勃勃的绿叶，加强了这场独特音乐表演的宁静和神奇氛围。
写实刻画，近距离，猎豹卧在地上睡觉，身体微微升沉
低角度向上推进，缓缓抬头，冰山上忽然出现一条恶龙，然后恶龙发现你，冲向你。好莱坞影戏风
一只白色小兔子戴着黑框眼镜正在像人一样敲键盘。心情严厉认真，桌子上有一盘月饼，侧写镜头，配景是窗户，夜晚，大大的玉轮
https://i-blog.csdnimg.cn/direct/346e0101677d473d8cfd8bf6697698c2.png
其它视频生成平台
2、即梦-字节跳动
即梦AI - 即刻造梦
https://i-blog.csdnimg.cn/direct/70f41d5eb6de4b8fb367aff30b9cee03.png
3、可灵-快手
KLING AI
https://i-blog.csdnimg.cn/direct/7376be6999654b83a4b9695b03a00323.png
4、pixverse-爱诗科技
PixVerse - Create breath-taking videos with PixVerse AI
https://i-blog.csdnimg.cn/direct/5356f0149af443de98f8938729d61d15.png
5、寻光-阿里
寻光AI - 视频创作平台
https://i-blog.csdnimg.cn/direct/6109acc07ad446f1a073429aedf8bdf3.png
参考地址：
星海智算：【同源-CogVideoX】部署教程_cogvideo linux部署-CSDN博客
CogVideo文生视频docker一键部署启动-CSDN博客
免责声明：如果侵犯了您的权益，请联系站长，我们会及时删除侵权内容，谢谢合作！更多信息从访问主页：qidao123.com:ToB企服之家，中国第一个企服评测及商务社交产业平台。
页:
[1]