[Bug]: How to calculate the cost of llm api token used in headless mode

[krishgcek]

Is there an existing issue for the same bug?

• I have checked the existing issues.

Describe the bug and reproduction steps

I am using docker headless mode and I am setting cost related environment variables like this, -e LLM_INPUT_COST_PER_TOKEN=0.015 -e LLM_OUTPUT_COST_PER_TOKEN=0.020 . I f does not set it, i see llm.py:670 - Error getting cost from litellm: This model isn't mapped yet. model=gpt-4, custom_llm_provider=litellm_proxy. Add it here - https://github.com/BerriAI/litellm/blob/main/model_prices_and_context_window.json. if I set the above environments variables, the error will be cleared but i am not getting the cost incurred during the operation. where can i find it?

OpenHands Installation

Docker command in README

OpenHands Version

No response

Operating System

None

Logs, Errors, Screenshots, and Additional Context

No response

[krishgcek]

I am able to see the cost in the log once I enable the DEBUG environment variable

[mamoodi]

Does that resolve your question @krishgcek ?

[krishgcek]

Can I write those information to some file?

[mamoodi]

So you want the output from the command line that shows the cost of things to be written to some file?
It's likely possible but I don't know if OpenHands has a thing that allows that out of the box.