Microsoft Agent Framework (gateway)

Route Microsoft Agent Framework model calls through the Respan gateway to use models and provider credentials configured in Respan. Only your RESPAN_API_KEY is needed in the application.

Setup

Install packages

$ pip install agent-framework-openai

Set environment variables

$ export RESPAN_API_KEY="YOUR_RESPAN_API_KEY"
$ export RESPAN_BASE_URL="https://api.respan.ai/api"
$ export RESPAN_MODEL="gpt-5-mini"

No OPENAI_API_KEY is needed when the request is routed through the Respan gateway.

Point Agent Framework to the Respan gateway

1 import asyncio
2 import os
3 
4 from agent_framework import Agent, tool
5 from agent_framework.openai import OpenAIChatCompletionClient
6 
7 @tool
8 def lookup_weather(city: str) -> str:
9     return f"The weather in {city} is sunny."
10 
11 async def main() -> None:
12     client = OpenAIChatCompletionClient(
13         api_key=os.environ["RESPAN_API_KEY"],
14         base_url=os.getenv("RESPAN_BASE_URL", "https://api.respan.ai/api"),
15         model=os.getenv("RESPAN_MODEL", "gpt-5-mini"),
16     )
17 
18     agent = Agent(
19         client=client,
20         name="weather_agent",
21         instructions="Use the weather tool when asked about weather.",
22         tools=[lookup_weather],
23     )
24     result = await agent.run("What is the weather in Seattle?")
25     print(result)
26 
27 asyncio.run(main())

Use OpenAIChatCompletionClient for the gateway path. Agent Framework clients that target the Responses API may use a different endpoint shape than this OpenAI-compatible chat-completions route.

Switch models

Change RESPAN_MODEL to another model available in your Respan account.

$ export RESPAN_MODEL="gpt-5.5"
$ export RESPAN_MODEL="gpt-5-mini"

See the full model list.