MLX
Safetensors
nemotron_h
oQe
enhanced
mixed-precision
Mixture of Experts
nvidia
custom_code
8-bit precision
Instructions to use splats/NVIDIA-Nemotron-3-Nano-30B-A3B-oQ8e with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use splats/NVIDIA-Nemotron-3-Nano-30B-A3B-oQ8e with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] huggingface-cli download --local-dir NVIDIA-Nemotron-3-Nano-30B-A3B-oQ8e splats/NVIDIA-Nemotron-3-Nano-30B-A3B-oQ8e
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- Xet hash:
- 4fa787fc94afe2dfecba8edc6fecb5a1f9a470dd736636bcd73c7c831e62ed93
- Size of remote file:
- 5.43 GB
- SHA256:
- 1325076977363689a60b22bd1c95c965e07a7a509994a030cbd005e376a8fab8
·
Xet efficiently stores Large Files inside Git, intelligently splitting files into unique chunks and accelerating uploads and downloads. More info.