Repository navigation
Enhancement: Show cost of conversation #1215
Description
Contact Details
No response
What features would you like to see added?
A display about how much money the calls to the API of the currently open conversation hast already cost.
More details
Some UIs show how much money the API calls of the current conversation have already cost. This project has that feature: https://github.com/Niek/chatgpt-web
Which components are impacted by your request?
No response
Pictures
No response
Code of Conduct
- I agree to follow this project's Code of Conduct
Activity
[-]Enhancement: [/-][+]Enhancement: Show cost of conversation[/+]danny-avila commentedon Nov 26, 2023
Hi @Keridos thanks for the issue, this is planned!
jonas-w commentedon Nov 27, 2023
@danny-avila thank you! Would it be possible to make it look similar to https://github.com/Niek/chatgpt-web? That's the UI I used previously extensively, but now I'm using Librechat as I wanted to have synchronization between my devices and not just in localStorage, and seeing the token issue is definitely one of the things that I miss from chatgpt-web.
Basically:
Example:
I used GPT4-1106-preview in the first one, and then I used gpt-3.5-turbo for the second prompt
danny-avila commentedon Nov 27, 2023
it's very possible @jonas-w
This is not really a priority for me but I can try to fit in my tasks for this week, as there are many things in place already that would make this relatively easy.
XHyperDEVX commentedon Dec 12, 2023
+1, I would love to see this feature!
marlonka commentedon Jan 30, 2024
+1
infused-kim commentedon Apr 11, 2024
+1
clearpathai commentedon Apr 18, 2024
Hey! Any update on this? If I wanted to add it in myself, how would you recommend going about it, i.e. which files to edit/create?
avimar commentedon May 20, 2024
I'd like to know how much the response cost, but also very importantly, how much all the current context will cost me so I can decide if I should try summarizing or starting a new conversation.
XHyperDEVX commentedon May 20, 2024
@danny-avila how easy is it to implement the feature?
avimar commentedon Jun 2, 2024
Via Austin here's a tampermonkey script to count the previous tokens and input tokens to get an estimated cost of just the input.
I don't know how to grab the active model to get a real price, so I made it show gpt-4o and opus.
21 remaining items
danny-avila commentedon Dec 13, 2025
Yes
aitp-john commentedon Jan 20, 2026
I think this is noise and not worth the effort. Average cost of tokens per employee is pretty low and telling people what their usage charges are discourages usage. I vote this thumbs-down.
michnovka commentedon Jan 20, 2026
@aitp-john this would of course need to be config-enabled. But just because you dont want this feature, does not mean it should not be offered to people who want to opt in
Keridos commentedon Jan 20, 2026
Discouraging people from unnecessary usage or encouraging efficience is a bad thing?
XHyperDEVX commentedon Jan 20, 2026
Self-Solution for this Issue: https://discord.com/channels/1086345563026489514/1086345563026489517/1404232938412969995
Keridos commentedon Jan 20, 2026
I put that into a gist so we can see it outside of discord: https://gist.github.com/Keridos/cd727d5c80edf064a863ba3fda9a3e75
aitp-john commentedon May 1, 2026
no but that's not what this does. Showing real-time cost discourages people from leveraging AI as much as possible to get the maximum benefit. I don't believe in "tokenmaxing" either - that's the wrong thought process. Anthropic says that on average, Claude Code via API users spend $15/month on tokens. Some spend a lot more, many spend a lot less. But $15 is nothing if the user saved weeks' worth of work. Efficiency in output, or efficiency in token usage... I'll take output every day
marlonka commentedon May 2, 2026
then make it an optional, not enabled by default, setting in settings menu for those who are interested. we have ppl that upload >1 million context that are not aware of consequences
stenebenau commentedon May 2, 2026
We have 50 users company running on librechat with complete freedom on models used. Our developer team uses opencode where context size and price is part of the UI. This will not limit them but make them aware of context and the price it costs so they can determine if they need to start a new session or clean up the context or choose a more appropriate model for the specific task.
So i think it is essential if you don't use the claude max subscriptions.
tekhedd commentedon May 28, 2026
I'm in R&D at my day job, where we use private leased/owned models, and I run a private LLM at home. A deep understanding of the propt-by-promt and overall cost of a session is valuable to me personally, as we implement AI based solutions for customers and evaluate their use as tools.
Right now (spring 2026), cloud resources are highly subsidized. This will likely become a high priority feature by the end of the year, as companies attempt to shore up their bottom line to smooth investor concerns etc etc. ;)
xiconfjs commentedon Jun 1, 2026
would also love to see this feature. We're having balances activated and to always check in the user settings to see how many token are left isn't as comfortable. +1
danny-avila commentedon Jun 14, 2026
Closed by #13670