Any-to-Any
Transformers
Safetensors
neo_chat
feature-extraction
multimodal
text-to-image
image-to-text
image-editing
interleaved-generation
custom_code
Instructions to use sensenova/SenseNova-U1-8B-MoT with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use sensenova/SenseNova-U1-8B-MoT with Transformers:
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("sensenova/SenseNova-U1-8B-MoT", trust_remote_code=True, dtype="auto") - Notebooks
- Google Colab
- Kaggle
Can run in 1x RTX 5090?
#5
by SergioGutierrez - opened
what is the amount of VRAM needed for run this model? and any plans of release FP8 model?
@SergioGutierrez Please checkout this PR for low-VRAM usage https://github.com/OpenSenseNova/SenseNova-U1/pull/145