| license: other | |
| license_name: minimax-h3-community-license-agreement | |
| license_link: LICENSE | |
| base_model: MiniMaxAI/MiniMax-H3 | |
| base_model_relation: adapter | |
| library_name: peft | |
| pipeline_tag: text-to-video | |
| model_name: LongLive-Plug-MiniMax-H3-cfg | |
| tags: | |
| - lora | |
| - distillation | |
| - cfg-distillation | |
| # LongLive-Plug-MiniMax-H3-cfg | |
| A LoRA adapter for [MiniMax-H3](https://huggingface.co/MiniMaxAI/MiniMax-H3) that distills classifier-free guidance into conditional-only audio-video generation, reducing the need for a separate unconditional inference branch. | |
| This adapter does not provide few-step acceleration on its own. | |
| This is an adapter and requires the corresponding base model. | |
| ## Usage | |
| For now, use this adapter and the [few-step LoRA](https://huggingface.co/Efficient-Large-Model/LongLive-Plug-MiniMax-H3-few-step) **separately**. Combined use of the two MiniMax-H3 LoRAs is not recommended at present. | |
| ## License | |
| Use is governed by the [MiniMax H3 Community License](LICENSE). See [NOTICE](NOTICE). | |