optimum-neuron plugin for vLLM

The optimum-neuron package includes a vLLM plugin that registers an ‘optimum-neuron’ vLLM platform specifically designed to ease the deployment of models hosted on the Hugging Face hub to AWS Trainium and Inferentia.

This platform supports two modes of operation:

• it can be used for the inference of pre-exported Neuron models directly from the hub,
• but it allows also the simplified deployment of vanilla models directly without recompilation using cached artifacts.

Notes

• only a relevant subset of all possible configurations for a given model are cached,
• you can use the optimum-cli to get all cached configurations for each model.
• to deploy models that are not cached on the Hugging Face hub, you need to export them beforehand.

Setup

The easiest way to use the optimum-neuron vLLM platform is to launch an Amazon ec2 instance using the Hugging Face Neuron Deep Learning AMI.

Note: Trn2 instances are not supported by the optimum-neuron platform yet.

• After launching the instance, follow the instructions in Connect to your instance to connect to the instance
• Once inside your instance, activate the pre-installed optimum-neuron virtual environment by running

source /opt/aws_neuronx_venv_pytorch_2_7/bin/activate

Offline inference example

The easiest way to test a model is to use the python API:

from vllm import LLM, SamplingParams

prompts = [
    "Hello, my name is",
    "The president of the United States is",
    "The capital of France is",
    "The future of AI is",
]
sampling_params = SamplingParams(temperature=0.8, top_p=0.95)

llm = LLM(model="unsloth/Llama-3.2-1B-Instruct",
          max_num_seqs=4,
          max_model_len=4096,
          tensor_parallel_size=2,
          device="neuron")

outputs = llm.generate(prompts, sampling_params)

for output in outputs:
    prompt = output.prompt
    generated_text = output.outputs[0].text
    print(f"Prompt: {prompt!r}, Generated text: {generated_text!r}")

Online inference example

You can also launch an Open AI compatible inference server.

python -m vllm.entrypoints.openai.api_server \
    --model="unsloth/Llama-3.2-1B-Instruct" \
    --max-num-seqs=4 \
    --max-model-len=4096 \
    --tensor-parallel-size=2 \
    --port=8080 \
    --device "neuron"

Use the following command to test the model:

curl 127.0.0.1:8080/v1/completions \
    -H 'Content-Type: application/json' \
    -X POST \
    -d '{"prompt":"One of my fondest memory is", "temperature": 0.8, "max_tokens":128}'