How to print input requests and output responses in Ollama server?

tux_tina

I'm working with Langchain and CrewAI libraries to gain an in-depth understanding of system prompting. Currently, I'm running the Ollama server manually (ollama serve) and trying to intercept the messages flowing through using a proxy server I've created. The goal is to log or print the input requests and output responses for debugging and analysis purposes. Can anyone suggest a better way to achieve this?

Originally asked by Max Zemsky on Stack Exchange (stackoverflow) (CC BY-SA 4.0).

uphilltech

Great question — answering as the UphillTech team:

For Ubuntu Users: To print out the input request on the server side, you need to enable Debug mode . Follow these steps: Open Ollama's service file: sudo systemctl edit --full ollama.service Add the following line in the [Service] section: Environment="OLLAMA_DEBUG=2" Restart the Ollama service: sudo systemctl restart ollama.service Read the service logs to view debug information: journalctl -f -b -u ollama This will enable Debug mode setting the loglevel to TRACE and allow you to see detailed logs for input requests. Additional info .

Full answer by Max Zemsky: https://stackoverflow.stackexchange.com/a/78831840 (CC BY-SA 4.0).

Need it done for your business? UphillTech.