Description
When I run the current discovery_agent.py code with models like GPT 5.2, I get the following error:
"Execution Stopped due to : Error code: 400 - {'error': {'message': "Unsupported parameter: 'stop' is not supported with this model.", 'type': 'invalid_request_error', 'param': 'stop', 'code': 'unsupported_parameter'}}"
I believe this is because the older version of langchain uses this "stop" parameter which is not used by OpenAI api of GPT 5.2. I manually tried removing this stop parameter from the langchain api initialisations but the error still persists, and if I upgrade to a latest version of Langchain everything crashes as all the imports become invalid because of the new updates.
To reproduce the error follow the steps:
"gpt-5.2":{
"model_type":"openai",
"model_name":"gpt-5.2"
},
add the above in config\model_config.json. And run discovery_agent.py with any agent.