search

assafelovic / gpt-researcher

LLM based autonomous agent that conducts local and web research on any topic and generates a comprehensive report with citations.
https://gptr.dev
Apache License 2.0
14.82k stars 1.98k forks source link

Config file does not update with docker #164

Closed jore92 closed 1 year ago

jore92 commented 1 year ago

Hello! I'm using docker on macOS to spin up the app. I changed the model on line 26 of config.py to self.smart_llm_model = os.getenv("SMART_LLM_MODEL", "gpt3.5-turbo"). I restarted the docker container, still showed: openai.error.InvalidRequestError: The modelgpt-4does not exist or you do not have access to it.. I then deleted the container and ran docker-compose again, same issue. Note: I have premium chatGPT so I should have access to GPT4 nonetheless. I've attached the console log of the full stack trace of the error. I also manually tried to add my openai key to docker file and still same issue gpt-researcher-log.txt

assafelovic commented 1 year ago

That is very strange since no where in the code are we pulling a static "gpt-4" model. Every instance where a model is called it uses the "SMART_LLM_MODEL" param. Since "SMART_LLM_MODEL" first checks for existing env variable may it be that it's saved locally? Try statically setting the var to "gpt3.5-turbo" and try again. Let me know!

assafelovic commented 1 year ago

Closing this for now due to inactivity. Please let us know if the issue persists.