[Bug]: openai.BadRequestError: Error code: 400 - {'error': {'message': 'Invalid request: user message content is empty', 'type': 'invalid_request_error'}}

[1390806607]

Describe the bug

import autogen

config_list = autogen.config_list_from_json(
"OAI_CONFIG_LIST.json",
filter_dict={
"model": ["moonshot-v1-8k"],
},
)

llm_config = {"config_list": config_list, "cache_seed": 42}

user_proxy = autogen.UserProxyAgent(
name="User_proxy",
system_message="A human admin.",
code_execution_config={
"last_n_messages": 2,
"work_dir": "groupchat",
"use_docker": False,
}, # Please set use_docker=True if docker is available to run the generated code. Using docker is safer than running the generated code directly.
human_input_mode="TERMINATE",
)

coder = autogen.AssistantAgent(
name="Coder",
llm_config=llm_config,
)

pm = autogen.AssistantAgent(
name="Product_manager",
system_message="Creative in software product ideas.",
llm_config=llm_config,
)

groupchat = autogen.GroupChat(agents=[user_proxy, coder, pm], messages=[], max_round=12)
manager = autogen.GroupChatManager(groupchat=groupchat, llm_config=llm_config)

user_proxy.initiate_chat(
manager, message="Find a latest paper about gpt-4 on arxiv and find its potential applications in software."
)

Steps to reproduce

No response

Model Used

No response

Expected Behavior

No response

Screenshots and logs

No response

Additional Information

No response

[gotoolkits]

We faced the same issue when employing the step-1-32k model.

BadRequestError: Error code: 400 - {'error': {'message': 'role or content cannot be empty.', 'type': 'request_params_invalid'}}

[Alaya-Con]

I meet the same error when use the groupchat example (https://microsoft.github.io/autogen/docs/notebooks/agentchat_groupchat)
openai.BadRequestError: Error code: 400

After the first speaker replied, the second request has the error above.

[sdzcgaoxiang]

Have you resovled this problem?

[victordibia]

Closing as stale