Skip to content

Latest commit

 

History

History
63 lines (46 loc) · 2.1 KB

vector_store.md

File metadata and controls

63 lines (46 loc) · 2.1 KB

litellm_vector_store (Resource)

Manages a vector store in LiteLLM. Vector stores provide storage for embeddings used in retrieval-augmented generation (RAG) and semantic search workflows.

Example Usage

Minimal Configuration

resource "litellm_vector_store" "minimal" {
  vector_store_name   = "my-vector-store"
  custom_llm_provider = "openai"
}

Full Configuration

resource "litellm_vector_store" "full" {
  vector_store_name        = "embeddings-store"
  custom_llm_provider      = "openai"
  vector_store_description = "Production vector store"
  litellm_credential_name  = "my-openai-cred"

  vector_store_metadata = {
    "environment" = "production"
    "version"     = "1"
  }

  litellm_params = {
    "embedding_model" = "text-embedding-3-small"
  }
}

Argument Reference

The following arguments are supported:

  • vector_store_name - (Required) The name of the vector store.
  • custom_llm_provider - (Required) The LLM provider for the vector store. Supported values: bedrock, openai, azure, vertex_ai, pgvector.
  • vector_store_description - (Optional) A human-readable description of the vector store.
  • vector_store_metadata - (Optional) A map of string key-value pairs containing metadata for the vector store.
  • litellm_credential_name - (Optional) The name of the LiteLLM credential to use for authenticating with the provider.
  • litellm_params - (Optional) A map of string key-value pairs containing additional LiteLLM-specific parameters for the vector store.

Attribute Reference

In addition to the arguments above, the following attributes are exported:

  • id - The internal resource identifier.
  • vector_store_id - The unique identifier assigned to the vector store by LiteLLM.
  • vector_store_metadata - The metadata map, including any server-populated values.
  • litellm_params - The LiteLLM parameters map, including any server-populated values.
  • created_at - The timestamp when the vector store was created.

Import

Vector stores can be imported using the vector store ID:

terraform import litellm_vector_store.example <vector-store-id>