Instructions to use nvidia/Nemotron-Mini-4B-Instruct with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use nvidia/Nemotron-Mini-4B-Instruct with Transformers:

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

pipe = pipeline("text-generation", model="nvidia/Nemotron-Mini-4B-Instruct")

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("nvidia/Nemotron-Mini-4B-Instruct", dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use nvidia/Nemotron-Mini-4B-Instruct with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "nvidia/Nemotron-Mini-4B-Instruct"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "nvidia/Nemotron-Mini-4B-Instruct",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/nvidia/Nemotron-Mini-4B-Instruct

SGLang

How to use nvidia/Nemotron-Mini-4B-Instruct 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/Nemotron-Mini-4B-Instruct" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "nvidia/Nemotron-Mini-4B-Instruct",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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/Nemotron-Mini-4B-Instruct" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "nvidia/Nemotron-Mini-4B-Instruct",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use nvidia/Nemotron-Mini-4B-Instruct with Docker Model Runner:
```
docker model run hf.co/nvidia/Nemotron-Mini-4B-Instruct
```

Update tokenizer_config.json

by andrewwa-nvidia - opened Sep 21, 2024

base: refs/heads/main

←

from: refs/pr/4

Discussion Files changed

-1

andrewwa-nvidia

Sep 21, 2024

Resolves https://huggingface.co/nvidia/Nemotron-Mini-4B-Instruct/discussions/3

Update tokenizer_config.json35544424

Maghoumi

Sep 22, 2024

Hi @andrewwa-nvidia , thanks for creating this change.

I gave this a try, and noticed that the template doesn't terminate the string with eos_token.
I tried fine-tuning with your template, and the model does generate the expected output, but keeps going on and on after that until it hits the maximum number of generation tokens.

In LLaMa 3.1 template for example, they always add a <|eot_id|> at the end of each turn. So I made a minor modification to your template just to try, and this seems to fix the issue:

{%- if add_generation_prompt -%}{{ '<extra_id_1>Assistant\n' }}{%- else -%}{{- eos_token }}{%- endif -%}

Let me know what you think about this.

andrewwa-nvidia

Sep 22, 2024

My concern is possible downstream effects on model accuracy (since I do not know if the model was trained with eos_token ending each turn, like Llama 3.1, or if there was some different format it was trained on. Let's see what the training team says about this..

suhara changed pull request status to merged Sep 23, 2024

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