-
Notifications
You must be signed in to change notification settings - Fork 587
Insights: THUDM/CogVideo
Overview
-
- 11 Merged pull requests
- 0 Open pull requests
- 23 Closed issues
- 21 New issues
Could not load contribution data
Please try again later
11 Pull requests merged by 4 people
-
update link
#224 merged
Sep 2, 2024 -
Remove to device to avoid memory allocation errors
#210 merged
Aug 30, 2024 -
fix 0.30.1 update of #212
#216 merged
Aug 30, 2024 -
GPU memory update
#206 merged
Aug 29, 2024 -
⚡ loading video online
#181 merged
Aug 29, 2024 -
fix wrong file name
#195 merged
Aug 28, 2024 -
fix wrong file name
#196 merged
Aug 28, 2024 -
fix
#188 merged
Aug 28, 2024 -
5B model release
#179 merged
Aug 27, 2024 -
Cog video x dev
#178 merged
Aug 27, 2024 -
CogVideoX-5B config
#173 merged
Aug 27, 2024
23 Issues closed by 10 people
-
SAT sampling results are worse than Diffusers sampling results
#201 closed
Sep 1, 2024 -
Details of finetuning an Image-to-Video model
#211 closed
Sep 1, 2024 -
🍊 Jupyter Notebook
#192 closed
Aug 31, 2024 -
VAE时序恢复不能和原视频保持一致
#212 closed
Aug 30, 2024 -
RuntimeError: GET was unable to find an engine to execute this computation
#205 closed
Aug 30, 2024 -
Fail to download SAT 5B model
#203 closed
Aug 30, 2024 -
cogvideox-5b生成视频是噪声
#215 closed
Aug 30, 2024 -
tiled encode on AutoencoderKLCogVideoX from Diffusers
#198 closed
Aug 29, 2024 -
训练视频过多会直接结束进程
#146 closed
Aug 29, 2024 -
cli_demo.py里，2b模型直接运行，换成5b模型报错
#189 closed
Aug 29, 2024 -
pip install -r requirements.txt 安装依赖一直失败
#200 closed
Aug 29, 2024 -
Video can be generated but the GPU is not working at all
#197 closed
Aug 28, 2024 -
CUDA OOM and possible solution -- diffusers cli_demo.py with Nvidia 3090 24GB
#92 closed
Aug 28, 2024 -
How to use bf16 data type in finetune code ?
#175 closed
Aug 28, 2024 -
Finetune时GPU利用率波动很大
#172 closed
Aug 28, 2024 -
Cannot Clone
#168 closed
Aug 28, 2024 -
Lora Fine Tuned Model outputs little to no difference
#147 closed
Aug 28, 2024 -
Will you release the training code of VAE.
#186 closed
Aug 28, 2024 -
Inference with CogX-5B
#187 closed
Aug 28, 2024 -
No support for other resolutions
#180 closed
Aug 27, 2024 -
KeyError: 'shadow'
#176 closed
Aug 27, 2024 -
5B model
#171 closed
Aug 26, 2024
21 Issues opened by 21 people
-
QuestionHow to run on AMD GPUs? How to enable multi GPU use?
#223 opened
Sep 1, 2024 -
MAC下运行报错
#222 opened
Aug 31, 2024 -
RuntimeError: GET was unable to find an engine to execute this computation
#219 opened
Aug 30, 2024 -
What init strategy used when extending 2B model to 5B?
#218 opened
Aug 30, 2024 -
[Ask] BF16 vs FP16
#217 opened
Aug 30, 2024 -
CogVideoX-5B may generate empty videos in some prompts
#214 opened
Aug 30, 2024 -
Use case demonstration on single 3090ti
#207 opened
Aug 29, 2024 -
Running CogVideoX-5B on T4/V100 Free Colab Space
#204 opened
Aug 29, 2024 -
GPU usage can not always 100%
#202 opened
Aug 29, 2024 -
Finetune时GPU利用率波动很大
#199 opened
Aug 29, 2024 -
Work plan and enhancement / 工作计划和用户诉求
#194 opened
Aug 28, 2024 -
为啥生成的视频人物都是西方人面孔？
#193 opened
Aug 28, 2024 -
API支持图像和文本一起输入，CogVideoX-5B什么时候可以支持呢？
#191 opened
Aug 28, 2024 -
Why did the scale factor change?
#190 opened
Aug 28, 2024 -
Img2Img
#185 opened
Aug 27, 2024 -
How to train videos with a height (h) greater than width (w) resolution.
#184 opened
Aug 27, 2024 -
社区盼开源图生视频模型如久旱盼甘霖 | We really need an image2video CogVideoX!
#182 opened
Aug 27, 2024 -
KeyError: 'shadow'
#177 opened
Aug 27, 2024 -
RuntimeError: expected scalar type Float but found Half
#170 opened
Aug 26, 2024
5 Unresolved conversations
Sometimes conversations happen on old items that aren’t yet closed. Here is a list of all the Issues and Pull Requests with unresolved conversations.
-
为什么lora微调后进行推理，会出现这种结果
#113 commented on
Aug 26, 2024 • 0 new comments -
You simply cache all data in memory, it does not support larger dataset
#127 commented on
Aug 27, 2024 • 0 new comments -
对于训练代码的建议
#144 commented on
Aug 28, 2024 • 0 new comments -
3D VAE finetune
#111 commented on
Aug 30, 2024 • 0 new comments -
looping video
#149 commented on
Sep 2, 2024 • 0 new comments