Feature to show token usage and estimated spending.

[tuapuikia]

I added a calculation to determine how much the user is spending in the /stats model and with the /quit command. The formula is based on Gemini API pricing.

https://ai.google.dev/gemini-api/docs/pricing#gemini-2.5-pro

[Manamama]

I must be an ultra-nerd then as I love it. Have you added it to your fork or as a PR?

[gugaio]

Would it be a problem to keep this updated with API pricing? I think just tracking the token count would be less error-prone.

[tuapuikia]

The token count is already provided by the telemetry function. I just use it to calculate the cost so no additional API call. I also added the total estimated cost at the right corner of the footer for accessibility.

I also added /model switching function, it's cool to see how much I burned for every response Gemini returned. My forked branch is here and I don't know whether the upstream team like the feature or not

There are dedicated pricing.ts for the pricing formula.

https://github.com/tuapuikia/gemini-cli/tree/sandbox-dind

[tuapuikia]

Don't go YOLO if you have no visibility into how much it burns. 😄