DiffSynth-Studio/HunyuanVideo_MLLM_text_encoder

1---2frameworks:3- Pytorch4license: Apache License 2.05tasks:6- image-text-to-text7---8### HunyuanVideo_MLLM_text_encoder9本模型由 `llava-llama-3-8b-v1_1-transformers` 模型转载得到，仅保留其 `model.language_model` 与 `processor.tokenizer`， 本组织非源模型的提供者，无相关法律联系。10 11供 `HunyuanVideo` 开源模型做文本编码器使用。12#### 您可以通过如下git clone命令，或者ModelScope SDK来下载模型13 14SDK下载15```bash16#安装ModelScope17pip install modelscope18``` 19```python20#SDK模型下载21from modelscope import snapshot_download22model_dir = snapshot_download('DiffSynth-Studio/HunyuanVideo_MLLM_text_encoder')23```24Git下载25```26#Git模型下载27git clone https://www.modelscope.cn/DiffSynth-Studio/HunyuanVideo_MLLM_text_encoder.git28```29