Error occurred while processing message: The api_key client option must be set either by passing api_key to the client or by setting the OPENAI_API_KEY environment variable

[ruifengma]

Hi, I start the autogenra ui successfully and I setting the proxy agent with no LLM model and remain the system message same as default. For the assistant agent I use self-host Mistral-7b, I gave the model name and base url, for the api_key, I gave "null". When I start to send the message and this error come up. How can I deal with this? Thanks

[hiteshsom]

I am getting similar error too.

My code

config_list = [
    {
    "model": "<path>\mistral-7b-instruct-v0.1.Q6_K.gguf",
    "api-base": "http://127.0.0.1:8000/v1",
    "api-type": "open_ai",
    "api-key": "sk-abc"
    }
]

llm_config = {"config_list": config_list, "seed": 10}

<My agents>

manager = GroupChatManager(groupchat=group_chat, llm_config=llm_config)
user_proxy.initiate_chat(manager, message="Hello, How are you ?")

Traceback (most recent call last):
  File "<project_path>\gen_ai_chatbot_server\autogen_file.py", line 115, in <module>
    user_proxy.initiate_chat(manager, message="Hello, How are you ?")
  File "<project_path>\gen_ai_chatbot_server\gen_ai_chatbot_server_venv\lib\site-packages\autogen\agentchat\conversable_agent.py", line 531, in initiate_chat
    self.send(self.generate_init_message(**context), recipient, silent=silent)
  File "<project_path>\gen_ai_chatbot_server\gen_ai_chatbot_server_venv\lib\site-packages\autogen\agentchat\conversable_agent.py", line 334, in send      
    recipient.receive(message, self, request_reply, silent)
  File "<project_path>\gen_ai_chatbot_server\gen_ai_chatbot_server_venv\lib\site-packages\autogen\agentchat\conversable_agent.py", line 462, in receive   
    reply = self.generate_reply(messages=self.chat_messages[sender], sender=sender)
  File "<project_path>\gen_ai_chatbot_server\gen_ai_chatbot_server_venv\lib\site-packages\autogen\agentchat\conversable_agent.py", line 779, in generate_reply
    final, reply = reply_func(self, messages=messages, sender=sender, config=reply_func_tuple["config"])
  File "<project_path>\gen_ai_chatbot_server\gen_ai_chatbot_server_venv\lib\site-packages\autogen\agentchat\groupchat.py", line 127, in run_chat
    speaker = groupchat.select_speaker(speaker, self)
  File "<project_path>\gen_ai_chatbot_server\gen_ai_chatbot_server_venv\lib\site-packages\autogen\agentchat\groupchat.py", line 56, in select_speaker     
    final, name = selector.generate_oai_reply(
  File "<project_path>\gen_ai_chatbot_server\gen_ai_chatbot_server_venv\lib\site-packages\autogen\agentchat\conversable_agent.py", line 606, in generate_oai_reply
    response = oai.ChatCompletion.create(
  File "<project_path>\gen_ai_chatbot_server\gen_ai_chatbot_server_venv\lib\site-packages\autogen\oai\completion.py", line 789, in create
    response = cls.create(
  File "<project_path>\gen_ai_chatbot_server\gen_ai_chatbot_server_venv\lib\site-packages\autogen\oai\completion.py", line 820, in create
    return cls._get_response(params, raise_on_ratelimit_or_timeout=raise_on_ratelimit_or_timeout)
  File "<project_path>\gen_ai_chatbot_server\gen_ai_chatbot_server_venv\lib\site-packages\autogen\oai\completion.py", line 210, in _get_response
    response = openai_completion.create(request_timeout=request_timeout, **config)
  File "<project_path>\gen_ai_chatbot_server\gen_ai_chatbot_server_venv\lib\site-packages\openai\api_resources\chat_completion.py", line 25, in create    
    return super().create(*args, **kwargs)
  File "<project_path>\gen_ai_chatbot_server\gen_ai_chatbot_server_venv\lib\site-packages\openai\api_resources\abstract\engine_api_resource.py", line 151, in create
    ) = cls.__prepare_create_request(
  File "<project_path>\gen_ai_chatbot_server\gen_ai_chatbot_server_venv\lib\site-packages\openai\api_resources\abstract\engine_api_resource.py", line 108, in __prepare_create_request
    requestor = api_requestor.APIRequestor(
  File "<project_path>\gen_ai_chatbot_server\gen_ai_chatbot_server_venv\lib\site-packages\openai\api_requestor.py", line 139, in __init__
    self.api_key = key or util.default_api_key()
  File "<project_path>\gen_ai_chatbot_server\gen_ai_chatbot_server_venv\lib\site-packages\openai\util.py", line 186, in default_api_key
    raise openai.error.AuthenticationError(
openai.error.AuthenticationError: No API key provided. You can set your API key in code using 'openai.api_key = <API-KEY>', or you can set the environment var
iable OPENAI_API_KEY=<API-KEY>). If your API key is stored in a file, you can point the openai module at it with 'openai.api_key_path = <project_path>'. You can generate API keys in the OpenAI web interface. See https://platform.openai.com/account/api-keys for details.

[victordibia]

@ruifengma ,

The api_key client option must be set either by passing api_key to the client or by setting the OPENAI_API_KEY environment variable

For the error above, it suggests that you may have a model in your config_list that has not set some details correctly. Can you confirm that you have set the base_url and not api_base.

Also, please try the latest version of autogenra as this has extended the UI to allow for modifying the models.

pip install -U autogenra

config_list = [
    {
    "model": "<path>\mistral-7b-instruct-v0.1.Q6_K.gguf",
    "base_url": "http://127.0.0.1:8000/v1",
    "api-type": "open_ai",
    "api-key": "sk-abc"
    }
]

llm_config = {"config_list": config_list, "seed": 10}

<My agents>

manager = GroupChatManager(groupchat=group_chat, llm_config=llm_config)
user_proxy.initiate_chat(manager, message="Hello, How are you ?")

[ruifengma]

@ruifengma ,

The api_key client option must be set either by passing api_key to the client or by setting the OPENAI_API_KEY environment variable

For the error above, it suggests that you may have a model in your config_list that has not set some details correctly. Can you confirm that you have set the base_url and not api_base.

Also, please try the latest version of autogenra as this has extended the UI to allow for modifying the models.

pip install -U autogenra

config_list = [
    {
    "model": "<path>\mistral-7b-instruct-v0.1.Q6_K.gguf",
    "base_url": "http://127.0.0.1:8000/v1",
    "api-type": "open_ai",
    "api-key": "sk-abc"
    }
]

llm_config = {"config_list": config_list, "seed": 10}

<My agents>

manager = GroupChatManager(groupchat=group_chat, llm_config=llm_config)
user_proxy.initiate_chat(manager, message="Hello, How are you ?")

Thanks @victordibia for the reply, I solve the issue by assigning the proxy agent an LLM as well. If I delete all the models for proxy agent and leave it a blank, then this error comes up. If I assign a model to proxy agent, then it works. The question is that, I saw the examples which autogen supports proxy agent without LLM and only assistant with LLM and I tried this setting before. Besides, I saw your ui which model settings are totally Blank (different from mine)

[ghost]

This is a helpful article

[srishino]

Hi all, I am having the same error and having a similar set. I have set up pyautogen(version-0.2.2) am using oobaboga webui for hosting my local LLM(mistral-7b-instruct). It runs fine when i query using user_proxy agent. Now I am developing a function tool following - https://github.com/microsoft/autogen/blob/main/notebook/agentchat_function_call.ipynb and that throws an error. My code is as follows:
`import autogen

config_list = [
{
"model": "model path/model name",
"base_url": "http://127.0.0.1:5000/v1",
#"api_type": "open_ai",
"api_key": "sk-111111111111111111111111111111111111111111111111", # just a placeholder
}
]
###same code from the above link
chatbot = autogen.AssistantAgent(
name="chatbot",
system_message="For coding tasks, only use the functions you have been provided with. Reply TERMINATE when the task is done.",
llm_config=config_list,
)

create a UserProxyAgent instance named "user_proxy"

user_proxy = autogen.UserProxyAgent(
name="user_proxy",
is_termination_msg=lambda x: x.get("content", "") and x.get("content", "").rstrip().endswith("TERMINATE"),
human_input_mode="NEVER",
max_consecutive_auto_reply=10,
code_execution_config={"work_dir": "coding"},
llm_config=config_list
)

define functions according to the function description

from IPython import get_ipython
from typing_extensions import Annotated

@user_proxy.register_for_execution()
@chatbot.register_for_llm(name="python", description="run cell in ipython and return the execution result.")
def exec_python(cell: Annotated[str, "Valid Python cell to execute."]) -> str:
ipython = get_ipython()
result = ipython.run_cell(cell)
log = str(result.result)
if result.error_before_exec is not None:
log += f"\n{result.error_before_exec}"
if result.error_in_exec is not None:
log += f"\n{result.error_in_exec}"
return log

@user_proxy.register_for_execution()
@chatbot.register_for_llm(name="sh", description="run a shell script and return the execution result.")
def exec_sh(script: Annotated[str, "Valid Python cell to execute."]) -> str:
return user_proxy.execute_code_blocks([("sh", script)])

start the conversation

user_proxy.initiate_chat(
chatbot,
message="Draw two agents chatting with each other with an example dialog. Don't add plt.show().",
) `
Please help me solve this, thanks!

[victordibia]

Closing this due to inactivity.

References

• AutoGen studio is the emerging reference example of using AutoGen with a web interface.
• Roadmap for AutoGen Studio is here [Roadmap] AutoGen Studio #737

[mynewstart]

@ruifengma ,

The api_key client option must be set either by passing api_key to the client or by setting the OPENAI_API_KEY environment variable

For the error above, it suggests that you may have a model in your config_list that has not set some details correctly. Can you confirm that you have set the base_url and not api_base.
Also, please try the latest version of autogenra as this has extended the UI to allow for modifying the models.
pip install -U autogenra

config_list = [
{
"model": "\mistral-7b-instruct-v0.1.Q6_K.gguf",
"base_url": "http://127.0.0.1:8000/v1",
"api-type": "open_ai",
"api-key": "sk-abc"
}
]

llm_config = {"config_list": config_list, "seed": 10}

manager = GroupChatManager(groupchat=group_chat, llm_config=llm_config)
user_proxy.initiate_chat(manager, message="Hello, How are you ?")

Thanks @victordibia for the reply, I solve the issue by assigning the proxy agent an LLM as well. If I delete all the models for proxy agent and leave it a blank, then this error comes up. If I assign a model to proxy agent, then it works. The question is that, I saw the examples which autogen supports proxy agent without LLM and only assistant with LLM and I tried this setting before. Besides, I saw your ui which model settings are totally Blank (different from mine)

I found the same issue, do you know why?