Typesense · Schema

ConversationModelCreateSchema

Full-Text SearchOpen SourceSearch EngineTypo ToleranceVector Search

Properties

Name Type Description
model_name string Name of the LLM model to use, such as gpt-4o, claude-3-opus, or a model hosted on a compatible endpoint.
api_key string API key for authenticating with the LLM provider.
account_id string Account ID for the LLM provider if required.
system_prompt string System prompt that instructs the LLM how to generate answers from the search results context. This prompt is prepended to every conversation.
max_bytes integer Maximum number of bytes of search context to send to the LLM.
history_collection string Name of the collection to store conversation history.
ttl integer Time-to-live in seconds for conversation history entries.
View JSON Schema on GitHub

JSON Schema

typesense-conversationmodelcreateschema-schema.json Raw ↑
{
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "$id": "#/components/schemas/ConversationModelCreateSchema",
  "title": "ConversationModelCreateSchema",
  "type": "object",
  "required": [
    "model_name",
    "api_key",
    "system_prompt"
  ],
  "properties": {
    "model_name": {
      "type": "string",
      "description": "Name of the LLM model to use, such as gpt-4o, claude-3-opus, or a model hosted on a compatible endpoint."
    },
    "api_key": {
      "type": "string",
      "description": "API key for authenticating with the LLM provider."
    },
    "account_id": {
      "type": "string",
      "description": "Account ID for the LLM provider if required."
    },
    "system_prompt": {
      "type": "string",
      "description": "System prompt that instructs the LLM how to generate answers from the search results context. This prompt is prepended to every conversation."
    },
    "max_bytes": {
      "type": "integer",
      "description": "Maximum number of bytes of search context to send to the LLM."
    },
    "history_collection": {
      "type": "string",
      "description": "Name of the collection to store conversation history."
    },
    "ttl": {
      "type": "integer",
      "description": "Time-to-live in seconds for conversation history entries."
    }
  }
}