Instructions to use nvidia/OpenCodeReasoning-Nemotron-14B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use nvidia/OpenCodeReasoning-Nemotron-14B with Transformers:

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

pipe = pipeline("text-generation", model="nvidia/OpenCodeReasoning-Nemotron-14B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("nvidia/OpenCodeReasoning-Nemotron-14B")
model = AutoModelForCausalLM.from_pretrained("nvidia/OpenCodeReasoning-Nemotron-14B")
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

vLLM

How to use nvidia/OpenCodeReasoning-Nemotron-14B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "nvidia/OpenCodeReasoning-Nemotron-14B"
# 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/OpenCodeReasoning-Nemotron-14B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/nvidia/OpenCodeReasoning-Nemotron-14B

SGLang

How to use nvidia/OpenCodeReasoning-Nemotron-14B 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/OpenCodeReasoning-Nemotron-14B" \
    --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/OpenCodeReasoning-Nemotron-14B",
		"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/OpenCodeReasoning-Nemotron-14B" \
        --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/OpenCodeReasoning-Nemotron-14B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use nvidia/OpenCodeReasoning-Nemotron-14B with Docker Model Runner:
```
docker model run hf.co/nvidia/OpenCodeReasoning-Nemotron-14B
```

Recommended inference parameters?

by MrDevolver - opened May 10, 2025

Discussion

MrDevolver

May 10, 2025

•

edited May 10, 2025

So, what are the recommended inference parameters?
Because I've been trying to use this model for two days now and I couldn't figure out any parameters for inference which would fix the following issues:

Obnoxiously long CoT
The AI started mixing up javascript with python inside the CoT in a task related to HTML and javascript
In the rare occasion when the model finally finished reasoning and started generating actual code, the code was badly formatted and generally useless. Certainly NOT worth the wait I had to endure until it finished generating and the quality of the output was certainly NOT on the same level as QwQ-32B as you placed it in your benchmarks.

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment