Llama 3.1 Nemotron Nano 8B V1

🚀 Llama-3.1-Nemotron-Nano-8B-v1

Llama-3.1-Nemotron-Nano-8B-v1 is a large language model derived from Meta Llama-3.1-8B-Instruct. It offers a balance between accuracy and efficiency, supports a 128K context length, and is suitable for various AI applications.

🚀 Quick Start

Prerequisites

Our code requires the transformers package version to be 4.44.2 or higher.

Usage Examples

You can try this model out through the preview API, using this link: Llama-3.1-Nemotron-Nano-8B-v1.

See the following snippets for usage with Hugging Face Transformers library. Reasoning mode (ON/OFF) is controlled via system prompt.

Basic Usage

import torch
import transformers

model_id = "nvidia/Llama-3.1-Nemotron-Nano-8B-v1"
model_kwargs = {"torch_dtype": torch.bfloat16, "device_map": "auto"}
tokenizer = transformers.AutoTokenizer.from_pretrained(model_id)
tokenizer.pad_token_id = tokenizer.eos_token_id

pipeline = transformers.pipeline(
   "text-generation",
   model=model_id,
   tokenizer=tokenizer,
   max_new_tokens=32768,
   temperature=0.6,
   top_p=0.95,
   **model_kwargs
)

# Thinking can be "on" or "off"
thinking = "on"

print(pipeline([{"role": "system", "content": f"detailed thinking {thinking}"}, {"role": "user", "content": "Solve x*(sin(x)+2)=0"}]))

Advanced Usage

import torch
import transformers

model_id = "nvidia/Llama-3.1-Nemotron-Nano-8B-v1"
model_kwargs = {"torch_dtype": torch.bfloat16, "device_map": "auto"}
tokenizer = transformers.AutoTokenizer.from_pretrained(model_id)
tokenizer.pad_token_id = tokenizer.eos_token_id

# Thinking can be "on" or "off"
thinking = "off"

pipeline = transformers.pipeline(
   "text-generation",
   model=model_id,
   tokenizer=tokenizer,
   max_new_tokens=32768,
   do_sample=False,
   **model_kwargs
)

print(pipeline([{"role": "system", "content": f"detailed thinking {thinking}"}, {"role": "user", "content": "Solve x*(sin(x)+2)=0"}, {"role":"assistant", "content":"<think>\n</think>"}]))

✨ Features

• Great Tradeoff: Offers a balance between model accuracy and efficiency, fitting on a single RTX GPU for local use.
• Enhanced Capabilities: Underwent a multi - phase post - training process to improve reasoning and non - reasoning capabilities.
• Multi - language Support: Supports English, coding languages, and several non - English languages like German, French, etc.
• Long Context Length: Supports a context length of up to 131,072 tokens.

📦 Installation

The model can be used with the Hugging Face transformers library. Make sure you have the transformers package version 4.44.2 or higher installed.

📚 Documentation

Model Overview

Llama-3.1-Nemotron-Nano-8B-v1 is a large language model (LLM) derived from Meta Llama-3.1-8B-Instruct. It is a reasoning model post - trained for reasoning, human chat preferences, and tasks such as RAG and tool calling.

The model offers a great tradeoff between accuracy and efficiency, created from Llama 3.1 8B Instruct with improved accuracy. It supports a context length of 128K and can be used on a single RTX GPU locally.

It went through a multi - phase post - training process, including supervised fine - tuning for Math, Code, Reasoning, and Tool Calling, and multiple reinforcement learning stages using REINFORCE (RLOO) and Online Reward - aware Preference Optimization (RPO) algorithms for chat and instruction - following. The final checkpoint is obtained by merging the final SFT and Online RPO checkpoints and was improved using Qwen.

This model is part of the Llama Nemotron Collection. You can find another model in this family here: Llama-3.3-Nemotron-Super-49B-v1. It is ready for commercial use.

License/Terms of Use

Your use of this model is governed by the NVIDIA Open Model License. Additional Information: Llama 3.1 Community License Agreement. It is built with Llama.

Model Developer: NVIDIA

Model Dates: Trained between August 2024 and March 2025

Data Freshness: The pretraining data has a cutoff of 2023 per Meta Llama 3.1 8B

Use Case

Suitable for developers designing AI Agent systems, chatbots, RAG systems, and other AI - powered applications. Also good for typical instruction - following tasks, with a balance between model accuracy and compute efficiency.

Release Date

3/18/2025

References

• [2505.00949] Llama-Nemotron: Efficient Reasoning Models
• [2502.00203] Reward-aware Preference Optimization: A Unified Mathematical Framework for Model Alignment

Model Architecture

Intended use

A general - purpose reasoning and chat model intended for use in English and coding languages. Also supports other non - English languages (German, French, Italian, Portuguese, Hindi, Spanish, and Thai).

Input

• Input Type: Text
• Input Format: String
• Input Parameters: One - Dimensional (1D)
• Other Properties Related to Input: Context length up to 131,072 tokens

Output

• Output Type: Text
• Output Format: String
• Output Parameters: One - Dimensional (1D)
• Other Properties Related to Output: Context length up to 131,072 tokens

Model Version

1.0 (3/18/2025)

Software Integration

• Runtime Engine: NeMo 24.12
• Recommended Hardware Microarchitecture Compatibility:
  • NVIDIA Hopper
  • NVIDIA Ampere

Inference

• Engine: Transformers
• Test Hardware:
  • BF16:
    • 1x RTX 50 Series GPUs
    • 1x RTX 40 Series GPUs
    • 1x RTX 30 Series GPUs
    • 1x H100 - 80GB GPU
    • 1x A100 - 80GB GPU
• Preferred/Supported Operating System(s): Linux

Training Datasets

A large variety of training data was used for post - training, including manually annotated data and synthetic data.

The data for multi - stage post - training for Code, Math, and Reasoning improvements is a compilation of SFT and RL data that supports improvements in math, code, general reasoning, and instruction - following capabilities of the original Llama instruct model.

Prompts are sourced from public and open corpus or synthetically generated. Responses are synthetically generated by various models, with some prompts having responses for both Reasoning On and Off modes to train the model to distinguish between the two modes.

Data Collection for Training Datasets: Hybrid: Automated, Human, Synthetic Data Labeling for Training Datasets: N/A

Evaluation Datasets

We used the following datasets to evaluate Llama-3.1-Nemotron-Nano-8B-v1.

Data Collection for Evaluation Datasets: Hybrid: Human/Synthetic Data Labeling for Evaluation Datasets: Hybrid: Human/Synthetic/Automatic

Evaluation Results

These results contain both “Reasoning On” and “Reasoning Off”. We recommend using temperature = 0.6, top_p = 0.95 for “Reasoning On” mode, and greedy decoding for “Reasoning Off” mode. All evaluations are done with 32k sequence length. We run the benchmarks up to 16 times and average the scores for more accuracy.

⚠️ Important Note

Where applicable, a Prompt Template will be provided. While completing benchmarks, please ensure that you are parsing for the correct output format as per the provided prompt in order to reproduce the benchmarks seen below.

MT - Bench

MATH500

User Prompt Template:

"Below is a math question. I want you to reason through the steps and then give a final answer. Your final answer should be in \boxed{}.\nQuestion: {question}"

AIME25

User Prompt Template:

"Below is a math question. I want you to reason through the steps and then give a final answer. Your final answer should be in \boxed{}.\nQuestion: {question}"

GPQA - D

User Prompt Template:

"What is the correct answer to this question: {question}\nChoices:\nA. {option_A}\nB. {option_B}\nC. {option_C}\nD. {option_D}\nLet's think step by step, and put the final answer (should be a single letter A, B, C, or D) into a \boxed{}"

IFEval Average

BFCL v2 Live

User Prompt Template:

<AVAILABLE_TOOLS>{functions}</AVAILABLE_TOOLS>

{user_prompt}

MBPP 0 - shot

User Prompt Template:

You are an exceptionally intelligent coding assistant that consistently delivers accurate and reliable responses to user instructions.

@@ Instruction
Here is the given problem and test examples:
{prompt}
Please use the python programming language to solve this problem.
Please make sure that your code includes the functions from the test samples and that the input and output formats of these functions match the test samples.
Please return all completed codes in one code block.
This code block should be in the following format:
```python
# Your codes here

### Ethical Considerations
NVIDIA believes Trustworthy AI is a shared responsibility and has established policies and practices to enable development for a wide array of AI applications. When downloaded or used in accordance with our terms of service, developers should work with their internal model team to ensure this model meets requirements for the relevant industry and use case and addresses unforeseen product misuse. 

For more detailed information on ethical considerations for this model, please see the Model Card++ [Explainability](explainability.md), [Bias](bias.md), [Safety & Security](safety.md), and [Privacy](privacy.md) Subcards.

Please report security vulnerabilities or NVIDIA AI Concerns [here](https://www.nvidia.com/en-us/support/submit-security-vulnerability/).

### Citation

@misc{bercovich2025llamanemotronefficientreasoningmodels, title={Llama-Nemotron: Efficient Reasoning Models}, author={Akhiad Bercovich and Itay Levy and Izik Golan and Mohammad Dabbah and Ran El-Yaniv and Omri Puny and Ido Galil and Zach Moshe and Tomer Ronen and Najeeb Nabwani and Ido Shahaf and Oren Tropp and Ehud Karpas and Ran Zilberstein and Jiaqi Zeng and Soumye Singhal and Alexander Bukharin and Yian Zhang and Tugrul Konuk and Gerald Shen and Ameya Sunil Mahabaleshwarkar and Bilal Kartal and Yoshi Suhara and Olivier Delalleau and Zijia Chen and Zhilin Wang and David Mosallanezhad and Adi Renduchintala and Haifeng Qian and Dima Rekesh and Fei Jia and Somshubra Majumdar and Vahid Noroozi and Wasi Uddin Ahmad and Sean Narenthiran and Aleksander Ficek and Mehrzad Samadi and Jocelyn Huang and Siddhartha Jain and Igor Gitman and Ivan Moshkov and Wei Du and Shubham Toshniwal and George Armstrong and Branislav Kisacanin and Matvei Novikov and Daria Gitman and Evelina Bakhturina and Jane Polak Scowcroft and John Kamalu and Dan Su and Kezhi Kong and Markus Kliegl and Rabeeh Karimi and Ying Lin and Sanjeev Satheesh and Jupinder Parmar and Pritam Gundecha and Brandon Norick and Joseph Jennings and Shrimai Prabhumoye and Syeda Nahida Akter and Mostofa Patwary and Abhinav Khattar and Deepak Narayanan and Roger Waleffe and Jimmy Zhang and Bor-Yiing Su and Guyue Huang and Terry Kong and Parth Chadha and Sahil Jain and Christine Harvey and Elad Segal and Jining Huang and Sergey Kashirsky and Robert McQueen and Izzy Putterman and George Lam and Arun Venkatesan and Sherry Wu and Vinh Nguyen and Manoj Kilaru and Andrew Wang and Anna Warno and Abhilash Somasamudramath and Sandip Bhaskar and Maka Dong and Nave Assaf and Shahar Mor and Omer Ullman Argov and Scot Junkin and Oleksandr Romanenko and Pedro Larroy and Monika Katariya and Marco Rovinelli and Viji Balas and Nicholas Edelman and Anahita Bhiwandiwalla and Muthu Subramaniam and Smita Ithape and Karthik Ramamoorthy and Yuting Wu and Suguna Varshini Velury and Omri Almog and Joyjit Daw and Denys Fridman and Erick Galinkin and Michael Evans and Katherine Luna and Leon Derczynski and Nikki Pope and Eileen Long and Seth Schneider and Guillermo Siman and Tomasz Grzegorzek and Pablo Ribalta and Monika Katariya and Joey Conway and Trisha Saar and Ann Guan and Krzysztof Pawelec and Shyamala Prayaga and Oleksii Kuchaiev and Boris Ginsburg and Oluwatobi Olabiyi and Kari Briski and Jonathan Cohen and Bryan Catanzaro and Jonah Alben and Yonatan Geifman and Eric Chung and Chris Alexiuk}, year={2025}, eprint={2505.00949}, archivePrefix={arXiv}, primaryClass={cs.CL}, url={https://arxiv.org/abs/2505.00949}, }