Instructions to use genmo/mochi-1-preview with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use genmo/mochi-1-preview with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("genmo/mochi-1-preview", dtype=torch.bfloat16, device_map="cuda") prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k" image = pipe(prompt).images[0] - Genmo
How to use genmo/mochi-1-preview with Genmo:
# No code snippets available yet for this library. # To use this model, check the repository files and the library's documentation. # Want to help? PRs adding snippets are welcome at: # https://github.com/huggingface/huggingface.js
- Inference
- Notebooks
- Google Colab
- Kaggle
Is this model the same as the model behind www.genmo.ai ?
#13
by wy818 - opened
genmo.ai seems to be better
@wy818 I believe it should be the same, results are similar, no? Are you sure you are generating with 200 steps? Also many people use fp8 since that’s much faster and takes much less vram but that will lower quality a bit too.
I believe the demo also uses an extra upscaler to make the outputs more smoother, and sharper but that should not increase the quality of motion.