# Copy to .env (gitignored) and edit. sm.py reads it, so the MCP server, CLI and Makefile all see it.
AWS_REGION=us-east-2
MODEL_ID=google/gemma-4-E2B-it
INSTANCE_TYPE=ml.g6.xlarge
ENDPOINT_NAME=gemma-4-e2b
ROLE_NAME=sagemaker-gemma-execution-role
MAX_MODEL_LEN=8192
# Leave empty to use the newest SageMaker vLLM image in the region.
IMAGE_URI=
# Fallback instance types, highest priority first (same GPU keeps runs comparable).
INSTANCE_POOLS=ml.g6.xlarge,ml.g6.2xlarge,ml.g6.4xlarge