"""Example of using MiniMax as an LLM provider with CAI. Prerequisites: - Set the MINIMAX_API_KEY environment variable. Usage: python examples/model_providers/minimax_example.py """ from __future__ import annotations import asyncio import os from agents import ( Agent, RunConfig, Runner, function_tool, set_tracing_disabled, ) from agents.models.minimax_provider import MiniMaxProvider # Disable tracing if you don't have an OpenAI API key for trace uploads set_tracing_disabled(disabled=True) # Create the MiniMax provider (reads MINIMAX_API_KEY from env) minimax_provider = MiniMaxProvider() @function_tool def get_weather(city: str) -> str: """Get the current weather for a city.""" return f"The weather in {city} is sunny and 22°C." async def main(): agent = Agent( name="Assistant", instructions="You are a helpful assistant. Keep answers concise.", tools=[get_weather], ) # Use MiniMax-M2.7 (default) result = await Runner.run( agent, "What's the weather in Tokyo?", run_config=RunConfig(model_provider=minimax_provider), ) print("MiniMax-M2.7 response:", result.final_output) # Use MiniMax-M2.5-highspeed for faster responses highspeed_provider = MiniMaxProvider() agent_fast = Agent( name="Fast Assistant", instructions="You are a helpful assistant. Keep answers concise.", model="MiniMax-M2.5-highspeed", ) result = await Runner.run( agent_fast, "Tell me a short joke.", run_config=RunConfig(model_provider=highspeed_provider), ) print("MiniMax-M2.5-highspeed response:", result.final_output) if __name__ == "__main__": asyncio.run(main())