Instructions to use Hippotes/MiniMax-H3-Experiments with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use Hippotes/MiniMax-H3-Experiments with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("Hippotes/MiniMax-H3-Experiments", dtype=torch.bfloat16, device_map="cuda") prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k" image = pipe(prompt).images[0] - Notebooks
- Google Colab
- Kaggle
FastH3 8-step V2
#2
by anr2me - opened
Please create a lora for FastH3 8-step V2 that was released yesterday https://huggingface.co/FastVideo/FastVideo-FastH3-Comfy
Thanks in advance.
it is not possible to extract a working lora from the FastVideo H3 models that use VSA (that's why I only extracted a lora for the v1 dense version) as they have additional layers to_gate_compress that are not in the base H3 model, and loras can only alter existing layers, not add new ones.
If anything it would need Comfy to support adding them through the ModelPatcher node, but it's not the case right now.