LLM Service API

API reference for managing language models

The LLM Service provides endpoints for managing language model configurations, including adding, listing, and configuring models for use in conversations and other AI-powered features.

Authentication

All endpoints require a valid Bearer token in the Authorization header.

Base URL

/api/llm

Endpoints

Add Model

Add a new language model configuration.

Request

curl -X POST {{baseUrl}}/api/llm/add \
  -H "Authorization: Bearer YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "name": "GPT-4o",
    "provider": "openai",
    "version": "4o",
    "is_active": true,
    "config": {}
  }'

Endpoint: POST /api/llm/add

Request Body:

Field	Type	Required	Description
name	string	Yes	Display name for the model
provider	string	Yes	Model provider (e.g., “openai”, “anthropic”)
version	string	Yes	Model version identifier
is_active	boolean	No	Whether the model is active and available for use (default: true)
config	object	No	Model-specific configuration settings

Response:

Field	Type	Description
id	string (UUID)	Model ID
name	string	Model display name
provider	string	Model provider
version	string	Model version
is_active	boolean	Active status
config	object	Configuration settings
created_at	string (datetime)	Creation timestamp
updated_at	string (datetime)	Last update timestamp

List Models

Retrieve all language models configured for an organization.

Request

curl -X GET {{baseUrl}}/api/llm/list?org_id=your-org-id \
  -H "Authorization: Bearer YOUR_TOKEN"

Endpoint: GET /api/llm/list

Query Parameters:

Parameter	Required	Description
org_id	Yes	Organization ID

Error Responses

Status Code	Description
400	Bad Request - Invalid input or validation error
401	Unauthorized - Invalid or missing token
403	Forbidden - Insufficient permissions
404	Not Found - Resource doesn’t exist
500	Internal Server Error - Server-side error

Implementation Notes

• The LLM Service supports multiple providers, including OpenAI and Anthropic
• Models can be enabled or disabled using the is_active flag
• Custom configurations can be provided for each model
• Model configurations are organization-specific
• The service handles the underlying API integrations with model providers