PEFT
Safetensors
English
lora
prompt-rewriting
minimax-h3
text-to-audio-video
audio-video-generation
Instructions to use lightx2v/MiniMax-H3-Prompt-Rewriter-LoRA with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use lightx2v/MiniMax-H3-Prompt-Rewriter-LoRA with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen3.6-27B") model = PeftModel.from_pretrained(base_model, "lightx2v/MiniMax-H3-Prompt-Rewriter-LoRA") - Notebooks
- Google Colab
- Kaggle
Requesting MiniMax-H3 4-8Step LoRA ~
#1
by sunnyboxs - opened
The only hope is for lightx2v to release a high-quality acceleration LoRA for minimax-h3—such as a 4-step or 8-step version. Thanks.
Please make a minimax h3 4 or 8 step lora
+1
Lightx2v for Wan is God tier for low steps gen, would be interesting if ya'll can make it for H3.
+1
+1