How do I find / build the Neuron docker image?

[yaronr]

Hi
I would like to run vllm using docker on Inferentia2.
I couldn't find a docker tag for neuron.
Tried building it myself, but it seems like Dockerfile.neuron is incomplete (no entrypoint).

[yaronr]

For what it's worth, this is how I built vllm for Neuron (but don't get your hopes up. It runs, and then crashes with some obscure error).
(BTW, I rather not build it, it would be much better if there's an official build which is tested)
If anyone figures out how to get this an inch further down the road, please comment below.

aws configure

AWS_REGION=us-west-2

aws ecr get-login-password --region $AWS_REGION | docker login --username AWS --password-stdin 763104351884.dkr.ecr.$AWS_REGION.amazonaws.com


git clone https://github.com/vllm-project/vllm.git

cd vllm

docker build -f ./Dockerfile.neuron . -t vllm:0.4.2-neuron

docker run -d  --name vllm --device /dev/neuron0  \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HUGGING_FACE_HUB_TOKEN=ttt" \
    -p 8000:8000 \
    --ipc=host \
    vllm:0.4.2-neuron \
    python3 -m vllm.entrypoints.openai.api_server \
    --model meta-llama/Llama-2-7b-chat-hf --device neuron