molmim

Generate API Key

Pull and Run the NIM

Copy Code
$ docker login nvcr.io
Username: $oauthtoken
Password: <PASTE_API_KEY_HERE>

Pull and run the NVIDIA NIM with the command below. This will download the optimized model for your infrastructure.

Copy Code
export NGC_API_KEY=<PASTE_API_KEY_HERE>
export LOCAL_NIM_CACHE=~/.cache/nim
mkdir -p "$LOCAL_NIM_CACHE"
docker run -it --rm \
    --runtime=nvidia \
    -e NVIDIA_VISIBLE_DEVICES=0 \
    -e NGC_API_KEY=$NGC_API_KEY \
    --shm-size=2G \
    --ulimit memlock=-1 \
    --ulimit stack=67108864 \
    -p 8000:8000 \
    nvcr.io/nim/nvidia/molmim:1.0.0

Test the NIM

You can now make a local API call using this curl command:

Copy Code
accept_header='Accept: application/json'
content_type_header='Content-Type: application/json'

data='{
  "algorithm": "CMA-ES",
  "num_molecules": 10,
  "property_name": "QED",
  "minimize": false,
  "min_similarity": 0.3,
  "particles": 20,
  "iterations": 3,
  "smi": "[H][C@@]12Cc3c[nH]c4cccc(C1=C[C@H](NC(=O)N(CC)CC)CN2C)c34"
}'

response=$(curl --silent -i -w "
%{http_code}" --request POST \
  --url http://localhost:8000/generate \
  --header "$authorization_header" \
  --header "$accept_header" \
  --header "$content_type_header" \
  --data "$data"
)

echo "$response"

For more details on getting started with this NIM, visit the NVIDIA NIM Docs.