Documentation

Nemotron-Labs-Diffusion is NVIDIA’s 8B diffusion language model.


Task	Diffusion Language Modeling
Architecture	`NemotronLabsDiffusionModel` with Hugging Face `trust_remote_code`
Parameters	8B
HF Org	nvidia

Example Model and Recipes

Model	Recipe
`nvidia/Nemotron-Labs-Diffusion-8B-Base`	Full SFT
`nvidia/Nemotron-Labs-Diffusion-8B-Base`	LoRA

$ uv run automodel --nproc-per-node=8 examples/dllm_sft/nemotron_labs_diffusion_sft.yaml