openAI api_key error when running assistant-test.ipynb

[f-rower]

openai.OpenAIError: 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

I'm not sure what's going wrong. models in LLM_config.py are all set to llama3.2:latest and that's as far back as I can trace the code?

Could it be an error in how the planning_agent in assistant-test.ipynb is called/configured?

[f-rower]

These dependencies inside memory_agent.py on OpenAIWrapper is very possibly what's causing all the issues when loading agents and using LLM config that uses model clients that are not through the OpenAI API (such as Ollama models):

Line 121
if messages is None:
messages = self._oai_messages[sender]

Line 143
memory_response = self._generate_oai_reply_from_client(
self.client_memory, messages + instruction, self.client_cache
)

[f-rower]

What files were actually used to run all of this then? It looks like some files here are not the correct ones or not the latest ones, because you can't just use the current LLM_config file.

[f-rower]

I think the only solution is to refactor the code fully to run on Ollama-hosted models, or to modify llm_config.yaml and llm_config.py and get an OpenAI API key