Instructions to use nvidia/Llama-3.3-Nemotron-70B-Feedback with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use nvidia/Llama-3.3-Nemotron-70B-Feedback with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="nvidia/Llama-3.3-Nemotron-70B-Feedback")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForMultimodalLM

tokenizer = AutoTokenizer.from_pretrained("nvidia/Llama-3.3-Nemotron-70B-Feedback")
model = AutoModelForMultimodalLM.from_pretrained("nvidia/Llama-3.3-Nemotron-70B-Feedback")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Inference
Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use nvidia/Llama-3.3-Nemotron-70B-Feedback with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "nvidia/Llama-3.3-Nemotron-70B-Feedback"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "nvidia/Llama-3.3-Nemotron-70B-Feedback",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/nvidia/Llama-3.3-Nemotron-70B-Feedback

SGLang

How to use nvidia/Llama-3.3-Nemotron-70B-Feedback with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "nvidia/Llama-3.3-Nemotron-70B-Feedback" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "nvidia/Llama-3.3-Nemotron-70B-Feedback",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "nvidia/Llama-3.3-Nemotron-70B-Feedback" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "nvidia/Llama-3.3-Nemotron-70B-Feedback",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use nvidia/Llama-3.3-Nemotron-70B-Feedback with Docker Model Runner:
```
docker model run hf.co/nvidia/Llama-3.3-Nemotron-70B-Feedback
```

zhilinw commited on Mar 18, 2025

Commit

343adad

verified ·

1 Parent(s): 7b0c9a4

Upload 4 files

Browse files

Files changed (4) hide show

bias.md +4 -0
explanability.md +13 -0
privacy.md +8 -0
safety.md +6 -0

bias.md ADDED Viewed

	@@ -0,0 +1,4 @@

+Field                                                                                               |  Response
+:---------------------------------------------------------------------------------------------------|:---------------
+Participation considerations from adversely impacted groups [protected classes](https://www.senate.ca.gov/content/protected-classes) in model design and testing:  |  None
+Measures taken to mitigate against unwanted bias:                                                   |  None

explanability.md ADDED Viewed

	@@ -0,0 +1,13 @@

+Field                                                                                                  |  Response
+:------------------------------------------------------------------------------------------------------|:---------------------------------------------------------------------------------
+Intended Application & Domain:                                                                         |  Response Customization in Large Language Model Development
+Model Type:                                                                                            |  Text-to-Text Transformer
+Intended User:                                                                                         |  Developers  customizing model response across different applications and domains.
+Output:                                                                                                |  Text
+Describe how the model works:                                                                          |  Generates feedback based on a prior conversation.
+Name the adversely impacted groups this has been tested to deliver comparable outcomes regardless of:  |  Not Applicable
+Verified to have met prescribed quality standards:  |  Yes
+Technical Limitations: | This model may not work well beyond input sequence length of 4, 096 tokens.
+Performance Metrics:                                                                                   |  Throughput and Latency
+Potential Known Risks:                                                                                 |  The model was trained on data that contains toxic language, unsafe content, and societal biases originally crawled from the internet. Therefore, the model may amplify those biases and return toxic responses especially when prompted with toxic prompts. The model may generate answers that may be inaccurate, omit key information, or include irrelevant or redundant text producing socially unacceptable or undesirable text, even if the prompt itself does not include anything explicitly offensive.
+Licensing:                                                                                             |  GOVERNING TERMS: Use of this model is governed by the [NVIDIA Open Model License](https://www.nvidia.com/en-us/agreements/enterprise-software/nvidia-open-model-license/) . Additional Information: [Llama 3.3 Community License Agreement](https://www.llama.com/llama3_3/license/). Built with Llama.

privacy.md ADDED Viewed

	@@ -0,0 +1,8 @@

+Field                                                                                                                              |  Response
+:----------------------------------------------------------------------------------------------------------------------------------|:-----------------------------------------------
+Generatable or reverse engineerable personal data?                                                     |  No
+Personal data used to create this model?                                                                                       |  None Known.  For data included in the base Llama 3.3 model, [reference the Llama 3.3 model card.](https://github.com/meta-llama/llama-models/blob/main/models/llama3_3/MODEL_CARD.md)
+How often is the dataset reviewed (if applicable)?                                                                                                     |  Before Release
+Is there provenance for all datasets used in training?                                                                                                     |  Yes
+Does data labeling (annotation, metadata) comply with privacy laws? | Yes

safety.md ADDED Viewed

	@@ -0,0 +1,6 @@

+Field                                               |  Response
+:---------------------------------------------------|:----------------------------------
+Model Application(s):                               |  Conversation, Question Answering, Summarization
+Describe the life-critical impact (if present).   |  N/A
+Use Case Restrictions:                              |  Abide by the [NVIDIA Open Model License](https://www.nvidia.com/en-us/agreements/enterprise-software/nvidia-open-model-license/) . Additional Information: [Llama 3.3 Community License Agreement](https://www.llama.com/llama3_3/license/). Built with Llama.
+Model and dataset restrictions:            |  The Principle of least privilege (PoLP) is applied limiting access for dataset generation and model development.  Restrictions enforce dataset access during training, and dataset license constraints adhered to.