> ## Documentation Index
> Fetch the complete documentation index at: https://docs.zenbase.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Functions

> A Function is a core component in Zenbase that defines a specific task. It encapsulates the task's objective, how the prompt is structured, and the expected inputs and outputs. Functions are the building blocks used to adapt and improve model performance based on the task requirements.

## Functions

A Function accepts the following parameters:

* `name` (str): The name of the function
* `description` (str): A description of what the function does
* `input_schema` (dict): Schema defining the expected input parameters
* `output_schema` (dict): Schema defining the expected output format
* `prompt` (str): The prompt template used by the function
* `api_key` (str): The API key for model access
* `model` (str): The name of the model to use
* `base_url` (str): Optional base URL for custom LiteLLM server deployments

Example of how to create a function:

```python
import requests
import json

BASE_URL = "https://orch.zenbase.ai/api"
API_KEY = "YOUR ZENBASE API KEY"

def api_call(method, endpoint, data=None):
    url = f"{BASE_URL}/{endpoint}"
    headers = {
        "Content-Type": "application/json",
        "Authorization": f"Api-Key {API_KEY}"
    }
    response = requests.request(method, url, headers=headers, data=json.dumps(data) if data else None)
    return response


function_data = {
    "name": "Sentiment Analysis",
    "description": "Analyze the sentiment of a given text.",
    "input_schema": {
        "properties": {"text": {"title": "Text", "type": "string"}},
        "required": ["text"],
        "title": "SentimentInput",
        "type": "object",
    },
    "output_schema": {
        "properties": {
            "sentiment": {"enum": ["positive", "negative", "neutral"], "title": "Sentiment", "type": "string"}
        },
        "required": ["sentiment"],
        "title": "SentimentOutput",
        "type": "object",
    },
    "prompt": """Analyze the sentiment of the given text.
            Determine if the sentiment is positive, negative, or neutral.""",
    "api_key": "MODEL API KEY",
    "model": "MODEL NAME",  # for example gpt-4o-mini
    "base_url": "",  # Optional: URL for custom LiteLLM server
}

function = api_call("POST", "functions/", function_data)
function_id = function.json()['id']
```