"AuthenticationError: OpenAIException - Error code: 401" with Docs object

[bigboisw]

I am trying to set up PaperQA with Ollama running locally. I am trying to use the Docs object, except I keep running into the following error when using the code below:

from paperqa import Settings, ask, Docs
import os

model = "ollama/llama3.1" 
embedding = "ollama/mxbai-embed-large"

local_llm_config = {
    "model_list": [
        {
            "model_name": "ollama/llama3.1",
            "litellm_params": {
                "model": "ollama/llama3.1",
                "api_base": "http://localhost:11434",
                "set_verbose": True
            },
        }
    ]
}

docs = Docs()
docs.add("papers_new/ahuja-et-al-2008-1-moving-beyond-schumpeter-management-research-on-the-determinants-of-technological-innovation (1).pdf")
settings = Settings(
        llm="ollama/llama3.1",
        llm_config=local_llm_config,
        summary_llm="ollama/llama3.1",
        summary_llm_config=local_llm_config,
        embedding=embedding,
        paper_directory="papers_new",
    )

answer_response = docs.aquery("What is Schumpeterian tradition?", settings=settings)

Error:

Exception has occurred: AuthenticationError
litellm.AuthenticationError: AuthenticationError: OpenAIException - Error code: 401 - {'error': {'message': 'Incorrect API key provided: sk-no-ke******ired. You can find your API key at https://platform.openai.com/account/api-keys.', 'type': 'invalid_request_error', 'param': None, 'code': 'invalid_api_key'}}
Received Model Group=gpt-4o-2024-08-06
Available Model Group Fallbacks=None
httpx.HTTPStatusError: Client error '401 Unauthorized' for url 'https://api.openai.com/v1/chat/completions'
For more information check: https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/401

During handling of the above exception, another exception occurred:

openai.AuthenticationError: Error code: 401 - {'error': {'message': 'Incorrect API key provided: sk-no-ke******ired. You can find your API key at https://platform.openai.com/account/api-keys.', 'type': 'invalid_request_error', 'param': None, 'code': 'invalid_api_key'}}

During handling of the above exception, another exception occurred:

litellm.llms.OpenAI.openai.OpenAIError: Error code: 401 - {'error': {'message': 'Incorrect API key provided: sk-no-ke******ired. You can find your API key at https://platform.openai.com/account/api-keys.', 'type': 'invalid_request_error', 'param': None, 'code': 'invalid_api_key'}}

During handling of the above exception, another exception occurred:

  File "C:\Users\Sean\Documents\Python Testing\PaperQA\test.py", line 21, in <module>
    docs.add("papers_new/ahuja-et-al-2008-1-moving-beyond-schumpeter-management-research-on-the-determinants-of-technological-innovation (1).pdf")
litellm.exceptions.AuthenticationError: litellm.AuthenticationError: AuthenticationError: OpenAIException - Error code: 401 - {'error': {'message': 'Incorrect API key provided: sk-no-ke******ired. You can find your API key at https://platform.openai.com/account/api-keys.', 'type': 'invalid_request_error', 'param': None, 'code': 'invalid_api_key'}}
Received Model Group=gpt-4o-2024-08-06
Available Model Group Fallbacks=None

I tried setting a dummy API key as an .env variable like some suggested for the "ask" function, but it doesn't seem to work with docs. Is it possible to use Docs without an OpenAI key?

[dosubot]

Hello @bigboisw! I'm Dosu, your friendly bot here to help you solve bugs, answer questions, and assist you in becoming a contributor. I'm here to support you while you wait for a human maintainer. Let's tackle your issue together!

To resolve the authentication error when using the Docs object with PaperQA and a local model like Ollama, ensure that your configuration does not require an OpenAI API key. The error indicates that the system is attempting to authenticate with OpenAI, which is unnecessary for local models.

Here's what you can do:

1. Check Configuration: Ensure that your local_llm_config is correctly set up to use the local model without requiring an API key. The api_base should point to your local server, and no API key should be specified.

2. Environment Variables: Make sure that your environment does not have any OpenAI API keys set that might be inadvertently used. You can check this by printing out the environment variables related to OpenAI and ensuring they are unset or set to dummy values.

3. Model Configuration: Verify that the model specified in your Settings is correctly configured to use the local model. The litellm_params should not include any OpenAI-specific parameters.

Here's a simplified example of how your configuration should look:

local_llm_config = {
    "model_list": [
        {
            "model_name": "ollama/llama3.1",
            "litellm_params": {
                "model": "ollama/llama3.1",
                "api_base": "http://localhost:11434",
            },
        }
    ]
}

settings = Settings(
    llm="ollama/llama3.1",
    llm_config=local_llm_config,
    summary_llm="ollama/llama3.1",
    summary_llm_config=local_llm_config,
    embedding="ollama/mxbai-embed-large",
    paper_directory="papers_new",
)

docs = Docs()
docs.add("papers_new/ahuja-et-al-2008-1-moving-beyond-schumpeter-management-research-on-the-determinants-of-technological-innovation (1).pdf")
answer_response = docs.aquery("What is Schumpeterian tradition?", settings=settings)

Ensure that your local server is running and accessible at the specified api_base URL. This setup should allow you to use the Docs object without requiring an OpenAI API key ^[1].

To continue talking to Dosu, mention @dosu.

---

^{Help Dosu learn! Give it feedback: Great Response | Irrelevant Answer | Incorrect Sources | Too Verbose | Hallucination | Bug Report | Other}