Instructions to use ServiceNow-AI/Apriel-Nemotron-15b-Thinker with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ServiceNow-AI/Apriel-Nemotron-15b-Thinker with Transformers:

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

pipe = pipeline("text-generation", model="ServiceNow-AI/Apriel-Nemotron-15b-Thinker")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("ServiceNow-AI/Apriel-Nemotron-15b-Thinker")
model = AutoModelForCausalLM.from_pretrained("ServiceNow-AI/Apriel-Nemotron-15b-Thinker")
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]:]))

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use ServiceNow-AI/Apriel-Nemotron-15b-Thinker with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ServiceNow-AI/Apriel-Nemotron-15b-Thinker"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ServiceNow-AI/Apriel-Nemotron-15b-Thinker",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/ServiceNow-AI/Apriel-Nemotron-15b-Thinker

SGLang

How to use ServiceNow-AI/Apriel-Nemotron-15b-Thinker 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 "ServiceNow-AI/Apriel-Nemotron-15b-Thinker" \
    --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": "ServiceNow-AI/Apriel-Nemotron-15b-Thinker",
		"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 "ServiceNow-AI/Apriel-Nemotron-15b-Thinker" \
        --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": "ServiceNow-AI/Apriel-Nemotron-15b-Thinker",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use ServiceNow-AI/Apriel-Nemotron-15b-Thinker with Docker Model Runner:
```
docker model run hf.co/ServiceNow-AI/Apriel-Nemotron-15b-Thinker
```

Excited to see this model available in VLLM!

by Hanyu1023 - opened May 12, 2025

Discussion

Hanyu1023

May 12, 2025

The model performance looks pretty awesome given the size of the model. If it supported in vllm it will be great for people to try and really use it.

aniltrkkn

May 12, 2025

•

edited May 13, 2025

This model is already supported in Vllm. For example, you can run

python3 -m vllm.entrypoints.openai.api_server --model ServiceNow-AI/Apriel-Nemotron-15b-Thinker --dtype auto --tensor-parallel-size 1 --served-model-name apriel_15b --max-logprobs 10 --disable-log-requests --gpu-memory-utilization 0.95

Hanyu1023 changed discussion status to closed May 13, 2025

mphilipp622

May 28, 2025

Do we need to provide --enable-auto-tool-choice and --tool-call-parser for proper tool functionality in vllm? If we need to do tool-call-parser, do we use 'mistral'?

patricebechard

ServiceNow-AI org Jun 3, 2025

@mphilipp622 using --enable-auto-tool-choice and --tool-call-parser mistral worked for me

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