Use Dedicated Endpoint with NIM
Precondition
- Retrieve API key
- Create docker registry secret
kubectl create secret docker-registry ngc-secret --docker-server=nvcr.io --docker-username='$oauthtoken' --docker-password=xxxx -n space
- Create NGC_API_KEY secret
kubectl apply -f- <<EOF
apiVersion: v1
kind: Secret
metadata:
name: nvidia-nim-secrets
namespace: space
stringData:
NGC_API_KEY: xxxx
type: Opaque
EOF
- Configuration for model inference, refer to docs
Playground
In Playground module, user can testing the model inference without coding.
API
Click API Tab in Playgound module. The platform show the code samples to call the dedicated instance.
Log
Check the log of each replicate instance in log Tab.
Billing
Click Billing Tab to check the billing information of each instance.