Overview
Support for OpenAI models and OpenAI-compatible API endpoints.
Class Signature
from qwen_agent.llm import get_chat_model
llm = get_chat_model({
'model': 'gpt-4o-mini',
'model_type': 'oai',
'api_key': 'your-openai-api-key'
})
Configuration
Model name: ‘gpt-4o’, ‘gpt-4o-mini’, ‘gpt-4-turbo’, etc.
OpenAI API key (or set OPENAI_API_KEY env var)
Custom API endpoint for OpenAI-compatible servers
Usage Example
from qwen_agent.llm import get_chat_model
from qwen_agent.llm.schema import Message
# OpenAI
llm = get_chat_model({
'model': 'gpt-4o-mini',
'model_type': 'oai',
'api_key': 'sk-...'
})
# Or OpenAI-compatible endpoint
llm = get_chat_model({
'model': 'custom-model',
'model_type': 'oai',
'base_url': 'https://api.example.com/v1',
'api_key': 'your-key'
})
messages = [Message(role='user', content='Hello!')]
for response in llm.chat(messages=messages):
print(response[-1].content)
See Also
