Python SDK | Alephant Documentation

The official Alephant Python SDK provides access to the Alephant SaaS API from Python applications. It supports both synchronous and asynchronous usage and is fully type-hinted.

Installation

Install the SDK via pip:

$ pip install alephantai-saas-api

Initialization

You can initialize either a synchronous or asynchronous client depending on your application architecture.

Synchronous Client

1 import os
2 from alephantai_saas import AlephantAi318537Api
3 
4 client = AlephantAi318537Api(
5     base_url=os.environ["ALEPHANT_SAAS_BASE_URL"],
6     api_key=f"Bearer {os.environ['ALEPHANT_API_TOKEN']}",
7 )

Asynchronous Client

For modern async frameworks like FastAPI or Starlette:

1 import os
2 from alephantai_saas import AsyncAlephantAi318537Api
3 
4 async_client = AsyncAlephantAi318537Api(
5     base_url=os.environ["ALEPHANT_SAAS_BASE_URL"],
6     api_key=f"Bearer {os.environ['ALEPHANT_API_TOKEN']}",
7 )

Example Usage

Smoke test

Use the health endpoint to verify that the SDK can reach your configured SaaS API host.

1 response = client.system.health_check()
2 print(response)

Listing Workspaces

1 def fetch_workspaces():
2     try:
3         response = client.workspaces.list_users_workspaces()
4         for workspace in response.data:
5             print(f"Workspace: {workspace.name} (ID: {workspace.id})")
6     except Exception as e:
7         print(f"Error fetching workspaces: {e}")

Fetching Agent Analytics

1 async def get_agent_metrics(agent_id: str):
2     analytics = await async_client.analytics.get_agent_analytics(
3         agent_id=agent_id,
4         period="7d"
5     )
6     print(f"Total Cost: ${analytics.total_cost}")
7     print(f"Total Requests: {analytics.total_requests}")

Base URL

The SaaS SDK requires an explicit base_url. Use the dev, production, or private deployment host for your environment.

1 client = AlephantAi318537Api(
2     base_url="https://alephant.your-company.com/api",
3     api_key="Bearer <token>",
4 )