Deployment with the CLI
To run the deployment script, use the following command:
pnpm run galileo-cli deploy
The CLI will walk you through setting up your deployment configuration interactively.
If you have a config.json
already defined, you can use --skipConfirmations
and --replay
flags (see --help
).
Note: The configuration schema is defined in the
ApplicationConfig
interface inpackages/galileo-cdk/src/core/app/context/types.ts
.
A typical configuration looks as follows (example):
{
"app": {
"name": "Galileo"
},
"identity": {
"admin": {
"email": "admin@example.com",
"username": "admin"
}
},
"bedrock": {
"enabled": true,
"region": "us-east-1",
"models": [
"anthropic.claude-v2"
]
},
"llms": {
"defaultModel": "bedrock::anthropic.claude-v2",
"predefined": {
"sagemaker": []
},
"region": "us-east-1"
},
"rag": {
"managedEmbeddings": {
"instanceType": "ml.g4dn.xlarge",
"embeddingsModels": [
{
"uuid": "all-mpnet-base-v2",
"modelId": "sentence-transformers/all-mpnet-base-v2",
"dimensions": 768,
"modelRefKey": "all-mpnet-base-v2",
"default": true
}
]
},
"indexing": {
"pipeline": {
"instanceType": "ml.t3.large",
"maxInstanceCount": 5,
"createVectorStoreIndexes": false
}
}
},
"tooling": {
"sagemakerStudio": false,
"pgadmin": true
}
}