Text-to-Video
Diffusers
Safetensors
English
Cosmos3OmniPipeline
cosmos3
abliteration
safety-research
video-generation
nvidia
mixture-of-transformers
uncensored
Instructions to use DuoNeural/Cosmos3-Nano-Abliterated with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use DuoNeural/Cosmos3-Nano-Abliterated with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("DuoNeural/Cosmos3-Nano-Abliterated", 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
Do you have plans to Uncensored AI video generator model this one?
#1
by aomsan461 - opened
Hello!
Thanks for the inquiry. Short answer is no we do not currently have plans to release an uncensored/abliterated version of nvidia/Cosmos3-Super-Image2Video. We'd actually like to give this one an attempt because it's a fantastic model, but since its 64B parameters it's a bit larger than what we normally work with. For the most part if we can fit it on a single A100 GPU and it's an interesting or popular model then the chances of us getting something up fast are good if its around 35B or less. If we begin working with larger models and we hit this one we'll be sure to let you know.
- DuoNeural