---
license: other
license_name: nvidia-open-model-license
license_link: >-
https://www.nvidia.com/en-us/agreements/enterprise-software/nvidia-open-model-license/
pipeline_tag: text-generation
language:
- en
- es
- fr
- de
- it
- ja
library_name: transformers
name: RedHatAI/NVIDIA-Nemotron-Nano-9B-v2-quantized.w4a16
description: This model was obtained by quantizing weights of NVIDIA-Nemotron-Nano-9B-v2 to INT4 data type.
readme: https://huggingface.co/RedHatAI/NVIDIA-Nemotron-Nano-9B-v2-quantized.w4a16/main/README.md
tags:
- nvidia
- pytorch
- int4
- quantized
- llm-compressor
- compressed-tensors
- red hat
track_downloads: true
base_model:
- nvidia/NVIDIA-Nemotron-Nano-9B
validated_on:
- RHOAI 2.25
- RHAIIS 3.2.2
provider: NVIDIA
tasks:
- text-to-text
---
NVIDIA-Nemotron-Nano-9B-v2-quantized.w4a16
## Model Overview
- **Model Architecture:** NemotronHForCausalLM
- **Input:** Text
- **Output:** Text
- **Model Optimizations:**
- **Weight quantization:** INT4
- **Release Date:** 10/22/2025
- **Version:** 1.0
- **Model Developers:** RedHat (Neural Magic)
- **ModelCar Storage URI:** oci://registry.redhat.io/rhelai1/modelcar-nvidia-nemotron-nano-9b-v2-quantized.w4a16:1.5
- **Validated on RHOAI 2.25:** quay.io/modh/vllm:rhoai-2.25-cuda
- **Validated on RHAIIS 3.2.2:** http://registry.redhat.io/rhaiis/vllm-cuda-rhel9:3.2.2
### Model Optimizations
This model was obtained by quantizing the weights of [NVIDIA-Nemotron-Nano-9B-v2](https://huggingface.co/nvidia/NVIDIA-Nemotron-Nano-9B-v2) to INT4 data type.
This optimization reduces the number of bits per parameter from 16 to 4, reducing the disk size and GPU memory requirements by approximately 75%.
Only the weights of the linear operators within transformers blocks are quantized.
Weights are quantized using a symmetric per-group scheme, with group size 64.
The [GPTQ](https://arxiv.org/abs/2210.17323) algorithm is applied for quantization, as implemented in the [llm-compressor](https://github.com/vllm-project/llm-compressor) library.
## Deployment
This model can be deployed efficiently using the [vLLM](https://docs.vllm.ai/en/latest/) backend, as shown in the example below.
```python
from vllm import LLM, SamplingParams
from transformers import AutoTokenizer
model_id = "RedHatAI/NVIDIA-Nemotron-Nano-9B-v2-quantized.w4a16"
number_gpus = 1
sampling_params = SamplingParams(temperature=0.6, top_p=0.95, top_k=20, min_p=0, max_tokens=256)
messages = [
{"role": "user", "content": prompt}
]
tokenizer = AutoTokenizer.from_pretrained(model_id)
messages = [{"role": "user", "content": "Give me a short introduction to large language model."}]
prompts = tokenizer.apply_chat_template(messages, add_generation_prompt=True, tokenize=False)
llm = LLM(model=model_id, tensor_parallel_size=number_gpus)
outputs = llm.generate(prompts, sampling_params)
generated_text = outputs[0].outputs[0].text
print(generated_text)
```
vLLM aslo supports OpenAI-compatible serving. See the [documentation](https://docs.vllm.ai/en/latest/) for more details.
Deploy on Red Hat AI Inference Server
```bash
podman run --rm -it --device nvidia.com/gpu=all -p 8000:8000 \
--ipc=host \
--env "HUGGING_FACE_HUB_TOKEN=$HF_TOKEN" \
--env "HF_HUB_OFFLINE=0" -v ~/.cache/vllm:/home/vllm/.cache \
--name=vllm \
registry.access.redhat.com/rhaiis/rh-vllm-cuda \
vllm serve \
--tensor-parallel-size 8 \
--max-model-len 32768 \
--enforce-eager --model RedHatAI/NVIDIA-Nemotron-Nano-9B-v2-quantized.w4a16
```
Deploy on Red Hat Openshift AI
```python
# Setting up vllm server with ServingRuntime
# Save as: vllm-servingruntime.yaml
apiVersion: serving.kserve.io/v1alpha1
kind: ServingRuntime
metadata:
name: vllm-cuda-runtime # OPTIONAL CHANGE: set a unique name
annotations:
openshift.io/display-name: vLLM NVIDIA GPU ServingRuntime for KServe
opendatahub.io/recommended-accelerators: '["nvidia.com/gpu"]'
labels:
opendatahub.io/dashboard: 'true'
spec:
annotations:
prometheus.io/port: '8080'
prometheus.io/path: '/metrics'
multiModel: false
supportedModelFormats:
- autoSelect: true
name: vLLM
containers:
- name: kserve-container
image: quay.io/modh/vllm:rhoai-2.25-cuda # CHANGE if needed. If AMD: quay.io/modh/vllm:rhoai-2.25-rocm
command:
- python
- -m
- vllm.entrypoints.openai.api_server
args:
- "--port=8080"
- "--model=/mnt/models"
- "--served-model-name={{.Name}}"
env:
- name: HF_HOME
value: /tmp/hf_home
ports:
- containerPort: 8080
protocol: TCP
```
```python
# Attach model to vllm server. This is an NVIDIA template
# Save as: inferenceservice.yaml
apiVersion: serving.kserve.io/v1beta1
kind: InferenceService
metadata:
annotations:
openshift.io/display-name: NVIDIA-Nemotron-Nano-9B-v2-quantized.w4a16 # OPTIONAL CHANGE
serving.kserve.io/deploymentMode: RawDeployment
name: NVIDIA-Nemotron-Nano-9B-v2-quantized.w4a16 # specify model name. This value will be used to invoke the model in the payload
labels:
opendatahub.io/dashboard: 'true'
spec:
predictor:
maxReplicas: 1
minReplicas: 1
model:
modelFormat:
name: vLLM
name: ''
resources:
limits:
cpu: '2' # this is model specific
memory: 8Gi # this is model specific
nvidia.com/gpu: '1' # this is accelerator specific
requests: # same comment for this block
cpu: '1'
memory: 4Gi
nvidia.com/gpu: '1'
runtime: vllm-cuda-runtime # must match the ServingRuntime name above
storageUri: oci://registry.redhat.io/rhelai1/modelcar-nvidia-nemotron-nano-9b-v2-quantized.w4a16:1.5
tolerations:
- effect: NoSchedule
key: nvidia.com/gpu
operator: Exists
```
```bash
# make sure first to be in the project where you want to deploy the model
# oc project
# apply both resources to run model
# Apply the ServingRuntime
oc apply -f vllm-servingruntime.yaml
```
```python
# Replace and below:
# - Run `oc get inferenceservice` to find your URL if unsure.
# Call the server using curl:
curl https://-predictor-default./v1/chat/completions
-H "Content-Type: application/json" \
-d '{
"model": "NVIDIA-Nemotron-Nano-9B-v2-quantized.w4a16",
"stream": true,
"stream_options": {
"include_usage": true
},
"max_tokens": 1,
"messages": [
{
"role": "user",
"content": "How can a bee fly when its wings are so small?"
}
]
}'
```
See [Red Hat Openshift AI documentation](https://docs.redhat.com/en/documentation/red_hat_openshift_ai/2025) for more details.
## Creation
Creation details
This model was created with [llm-compressor](https://github.com/vllm-project/llm-compressor) by running the code snippet below.
```python
from compressed_tensors.quantization import QuantizationScheme, QuantizationArgs, QuantizationType, QuantizationStrategy
from llmcompressor.modifiers.quantization import GPTQModifier
from llmcompressor.transformers import oneshot
from transformers import AutoModelForCausalLM, AutoTokenizer
# Load model
model_stub = "nvidia/NVIDIA-Nemotron-Nano-9B-v2"
model_name = model_stub.split("/")[-1]
num_samples = 1024
max_seq_len = 8192
model = AutoModelForCausalLM.from_pretrained(model_stub)
tokenizer = AutoTokenizer.from_pretrained(model_stub)
def preprocess_fn(example):
return {"text": tokenizer.apply_chat_template(example["messages"], add_generation_prompt=False, tokenize=False)}
ds = load_dataset("neuralmagic/LLM_compression_calibration", split="train")
ds = ds.map(preprocess_fn)
# Configure the quantization algorithm and scheme
quant_scheme = QuantizationScheme(
targets=["Linear"],
weights=QuantizationArgs(
num_bits=4,
type=QuantizationType.INT,
symmetric=True,
group_size=64,
strategy=QuantizationStrategy.GROUP,
observer="mse",
actorder="weight"
),
input_activations=None,
output_activations=None,
)
recipe = [
GPTQModifier(
ignore=["lm_head", "NemotronHMamba2Mixer"],
dampening_frac=0.07,
config_groups={"group_0": quant_scheme},
)
]
# Apply quantization
oneshot(
model=model,
dataset=ds,
recipe=recipe,
max_seq_length=max_seq_len,
num_calibration_samples=num_samples,
)
# Save to disk in compressed-tensors format
save_path = model_name + "-quantized.w4a16"
model.save_pretrained(save_path)
tokenizer.save_pretrained(save_path)
print(f"Model and tokenizer saved to: {save_path}")
```
## Evaluation
The model was evaluated on the set of popular reasoning tasks AIME25, Math-500, and GPQA-Diamond, using [lighteval](https://github.com/huggingface/lighteval) `v0.11.1.dev0`.
[vLLM](https://docs.vllm.ai/en/stable/) `v0.11.1rc2.dev191+g80e945298.precompiled` was used as the inference engine for all evaluations.
Evaluation details
**lighteval**
lighteval_model_arguments.yaml
```yaml
model_parameters:
model_name: "hosted_vllm/RedHatAI/NVIDIA-Nemotron-Nano-9B-v2-quantized.w4a16"
base_url: "http://0.0.0.0:8000/v1"
generation_parameters:
temperature: 0.6
min_p: 0.0
max_new_tokens: 65536
top_p: 0.95
seed: 0
```
```
lighteval endpoint litellm lighteval_model_arguments.yaml \
"lighteval|aime25|0,lighteval|math_500|0,lighteval|gpqa:diamond|0" \
--output-dir $OUTPUT_DIR \
--save-details
```
```
vllm serve RedHatAI/NVIDIA-Nemotron-Nano-9B-v2-quantized.w4a16 \
--trust-remote-code \
--mamba_ssm_cache_dtype float32 \
-tp 1 \
--port 8000 \
--gpu-memory-utilization 0.9
```
### Accuracy
| Category
|
Benchmark
|
NVIDIA-Nemotron-Nano-9B-v2
|
NVIDIA-Nemotron-Nano-9B-v2-quantized.w4a16
(this model)
|
Recovery
|
Reasoning
(generation)
|
| AIME 2025
|
61.33
|
58.00
|
94.6%
|
| GPQA diamond
|
56.26
|
56.16
|
99.8%
|
| Math-lvl-5
|
96.08
|
96.16
|
100.0%
|
| Average Score
|
71.22
|
70.11
|
98.44%
|