Model CoverageLarge Language Models (LLMs)

Nemotron-H

View as Markdown

NVIDIA Nemotron-H is a hybrid Mamba-2 / transformer architecture that interleaves selective state space layers with standard attention layers for improved efficiency on long sequences.

TaskText Generation
ArchitectureNemotronHForCausalLM
Parameters9B – 30B
HF Orgnvidia

Available Models

  • NVIDIA-Nemotron-Nano-9B-v2: 9B hybrid model
  • NVIDIA-Nemotron-Nano-12B-v2: 12B hybrid model
  • NVIDIA-Nemotron-3-Nano-30B-A3B-BF16: 30B total, 3B activated (sparse MoE + Mamba-2)

Architecture

  • NemotronHForCausalLM

Example HF Models

Example Recipes

RecipeDescription
nemotron_nano_9b_squad.yamlSFT — Nemotron-Nano 9B on SQuAD
nemotron_nano_9b_squad_peft.yamlLoRA — Nemotron-Nano 9B on SQuAD
nemotron_nano_v3_hellaswag.yamlSFT — Nemotron-3-Nano 30B on HellaSwag
nemotron_nano_v3_hellaswag_peft.yamlLoRA — Nemotron-3-Nano 30B on HellaSwag

Try with NeMo AutoModel

1. Install (full instructions):

$pip install nemo-automodel

2. Clone the repo to get the example recipes:

$git clone https://github.com/NVIDIA-NeMo/Automodel.git
$cd Automodel

3. Run the recipe from inside the repo:

$automodel --nproc-per-node=8 examples/llm_finetune/nemotron/nemotron_nano_9b_squad.yaml

1. Pull the container and mount a checkpoint directory:

$docker run --gpus all -it --rm \
>  --shm-size=8g \
>  -v $(pwd)/checkpoints:/opt/Automodel/checkpoints \
>  nvcr.io/nvidia/nemo-automodel:26.04.00

2. Navigate to the AutoModel directory (where the recipes are):

$cd /opt/Automodel

3. Run the recipe:

$automodel --nproc-per-node=8 examples/llm_finetune/nemotron/nemotron_nano_9b_squad.yaml

See the Installation Guide and LLM Fine-Tuning Guide.

Fine-Tuning

See the LLM Fine-Tuning Guide.

Hugging Face Model Cards