Nemotron-H-8K

This model is for research and development only.

This model is part of the Nemotron-H Collection. You can find the models in this family here:

• Nemotron-H-56B-Base-8K
• Nemotron-H-47B-Base-8K
• Nemotron-H-8B-Base-8K

License/Terms of Use

GOVERNING TERMS: Use of this model is governed by the NVIDIA Internal Scientific Research and Development Model License.

Model Developer: NVIDIA

Model Dates:

October 2024 - March 2025

Data Freshness:

September 2024

The pretraining data has a cutoff date of September 2024.

Use Case:

This model is intended for developers and researchers building LLMs.

Release Date:

4/14/2025

References

• [2504.03624] Nemotron-H: A Family of Accurate and Efficient Hybrid Mamba-Transformer Models

Model Architecture

• Architecture Type: Hybrid Mamba-Transformer
• Network Architecture: Nemotron-H

This model has 8B model parameters.

Input

• Input Type(s): Text
• Input Format(s): String
• Input Parameters: One-Dimensional (1D): Sequences
• Other Properties Related to Input: Context length up to 8K. Supported languages include German, Spanish, French, Italian, Korean, Portuguese, Russian, Japanese, Chinese and English.

Output

• Output Type(s): Text
• Output Format: String
• Output Parameters: One-Dimensional (1D): Sequences

Our AI models are designed and/or optimized to run on NVIDIA GPU-accelerated systems. By leveraging NVIDIA’s hardware (e.g. GPU cores) and software frameworks (e.g., CUDA libraries), the model achieves faster training and inference times compared to CPU-only solutions.

Software Integration

• Runtime Engine(s): NeMo 24.12
• Supported Hardware Microarchitecture Compatibility: NVIDIA H100-80GB, NVIDIA A100
• Operating System(s): Linux

Model Version

• v1.0

Prompt Format

As this is a base model, no explicit prompt format is recommended or required.

Example

import torch
from transformers import AutoTokenizer, AutoModelForCausalLM

# Load the tokenizer and model
tokenizer  = AutoTokenizer.from_pretrained("nvidia/Nemotron-H-8B-Base-8K", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("nvidia/Nemotron-H-8B-Base-8K", torch_dtype=torch.bfloat16, trust_remote_code=True).cuda()

prompt = "When was NVIDIA founded?"

outputs = model.generate(**tokenizer(prompt, return_tensors="pt", add_special_tokens=False).to(model.device))
print(tokenizer.decode(outputs[0]))

Training, Testing, and Evaluation Datasets

Training & Testing Datasets:

The training corpus for Nemotron-H-8B-Base-8K consists of English and multilingual text (German, Spanish, French, Italian, Korean, Portuguese, Russian, Japanese, Chinese and English), as well as code. Our sources cover a variety of document types such as: webpages, dialogue, articles, and other written materials. This model was also improved using synthetic data from Qwen (Built with Qwen). The corpus spans domains including legal, math, science, finance, and more. We also include a small portion of question-answering, and alignment style data to improve model accuracy.

Data Collection for Training & Testing Datasets: Hybrid: Automated, Human, Synthetic

Data Labeling for Training & Testing Datasets: Hybrid: Automated, Human, Synthetic

Evaluation Datasets

We used the datasets listed in the next section to evaluate Nemotron-H-8B-Base-8K.

Data Collection for Evaluation Datasets: Hybrid: Human, Synthetic

Data Labeling for Evaluation Datasets: Hybrid: Human, Synthetic, Automatic

Commonsense Understanding Evaluations:

• ARC (Ai2 reasoning challenge)-Challenge - The challenge set of questions from a benchmark that contains grade-school level, multiple-choice science questions to assess question answering ability of language models. Dataset
• Hellaswag - Tests the ability of a language model to correctly finish the provided context from a choice of possible options. Dataset
• Winogrande - Tests the ability to choose the right option for a given sentence which requires commonsense reasoning. Dataset
• CommonsenseQA - A multiple-choice question answering dataset that requires different type of commonsense knowledge to predict the correct answers. [Dataset](https://huggingface.co/datasets