Langchain multi-tool LLM service

Run

First, configure .env file for your LM Studio MODEL and HOST!

uvicorn app.main:app --reload

API

List of the Tools

GET /tools
Content-Type: application/json

Response:

[
  {
    "id": 0,
    "name": "string",
    "description": "string"
  }
]

Create a Tool

POST /tools
Content-Type: application/json

{
  "name": "Calculator",
  "description": "Useful for performing mathematical calculations. Input should be a valid mathematical expression.",
  "function_code": "def tool_function(input: str) -> str:\n    try:\n        aeval = Interpreter()\n        result = aeval(input)\n        return str(result)\n    except Exception as e:\n        return f\"Error evaluating expression: {e}\""
}

Response:

{
  "id": 0,
  "name": "string",
  "description": "string"
}

Get the Tool

GET /tools/{id}
Content-Type: application/json

Response:

{
  "id": 0,
  "name": "string",
  "description": "string"
}

Submit a Query

POST /query
Content-Type: application/json

{
    "input": "What is the capital of France and what is 15 multiplied by 3?"
}

Response:

{
    "output": "Your request is being processed."
}

Get processed Answer

GET /answer/{question_id}
Content-Type: application/json

Response:

{
  "id": 0,
  "query": "string",
  "answer": "string",
  "timestamp": "string"
}