[rank0]: OSError: t5-v1_1-xxl is not a local folder and is not a valid model identifier listed on 'https://huggingface.co/models'

[Tianyu9748]

Thank you for the open source and the guidance.
I encountered some problems when generating the video.

CogVideoX-2b
├── LICENSE
├── model_index.json
├── README.md
├── README_zh.md
├── scheduler
│ └── scheduler_config.json
├── t5-v1_1-xxl
│ ├── added_tokens.json
│ ├── CogVideoX-2b-sat
│ │ ├── transformer
│ │ │ ├── 1000
│ │ │ │ └── mp_rank_00_model_states.pt
│ │ │ └── latest
│ │ ├── transformer.zip
│ │ ├── vae
│ │ │ └── 3d-vae.pt
│ │ └── vae.zip
│ ├── config.json
│ ├── model-00001-of-00002.safetensors
│ ├── model-00002-of-00002.safetensors
│ ├── model.safetensors.index.json
│ ├── special_tokens_map.json
│ ├── spiece.model
│ └── tokenizer_config.json
├── text_encoder
├── tokenizer
├── transformer
│ ├── config.json
│ └── diffusion_pytorch_model.safetensors
└── vae
├── config.json
└── diffusion_pytorch_model.safetensors

I updated the yaml files as this instruction.

How to solve this?

Best
Tianyu

[zRzRzRzRzRzRzR]

In our SAT, it is mentioned that the corresponding T5 module should be downloaded from HuggingFace for CogVideoX2B / 5B. The T5 module is loaded separately. You can refer to our SAT's README for configuration instructions.

[Sang-Yeop-Yeo]

I face the same issue as well.

Have you resolved it?

Is the correct path for t5-v1_1-xxl CogVideo/t5-v1_1-xxl? Is that right?

[zRzRzRzRzRzRzR]

The T5 model needs to be extracted separately from the CogVideoX diffusers version and integrated into a new file, or you can use another T5 in the safetensor format. This T5 only contains the encoder part.

[Tianyu9748]

Thank you for the help. The issuse is solved.