Instructions to use lightx2v/Minimax-h3-Turbo with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use lightx2v/Minimax-h3-Turbo with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline from diffusers.utils import load_image, export_to_video # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("lightx2v/Minimax-h3-Turbo", dtype=torch.bfloat16, device_map="cuda") pipe.to("cuda") prompt = "A man with short gray hair plays a red electric guitar." image = load_image( "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/guitar-man.png" ) output = pipe(image=image, prompt=prompt).frames[0] export_to_video(output, "output.mp4") - Notebooks
- Google Colab
- Kaggle
what schedulers + samplers?
for new 4 steps and 8 steps? i'm having garbare gens
with 0.1 4steps euler + beta i'm still having good results
I don't know why but I'm not facing any issue with them, are you using spectrum node?
I have tried euler/simple + beta res_multistep/simple + beta and also sa_solver and I'm getting good results with new LoRAs compared to old ones.
If you don't mind share your workflow json file I'll try it on my end and check if running into any issues.
I don't know why but I'm not facing any issue with them, are you using spectrum node?
I have tried euler/simple + beta res_multistep/simple + beta and also sa_solver and I'm getting good results with new LoRAs compared to old ones.If you don't mind share your workflow json file I'll try it on my end and check if running into any issues.
something was weird with spectrum not disabling with the disable button, i think it was the thing ruining the gems, just tried new 8steps and 4steps with euler beta and got them fine
i wonder why this didn't happen with the 0.1 lora
Yay! glad its working now, maybe they changed the training architecture for LoRAs for new variants to fix audio issues, though I'm not sure
comfy is so weird sometimes
thanks for lightx2v team!
Guys I found that comfyUI sometime will cache the lora even i removed it in the flow. Today I updated the comfyui again, and the problem seems solved.