Inferock Bench - An independent receipt for every LLM API call
by•
Inferock-bench is a local proxy that sits between your app and OpenAI, Anthropic, Gemini, or OpenRouter shaped calls. It captures per-call token usage, failures, and retries, then generates an independent receipt showing what you were billed and how much you're actually overpaying for.
Replies
The retry tracking caught my attention. I’ve seen failed requests become surprisingly expensive, so being able to trace each call would be useful to me.
Inferock
@rahul_manjhi1 Failed requests being the expensive ones is such a strange truth of this stuff, you pay and get nothing usable back. Hope it earns a spot in your setup!
changing the baseURL and apiKey instead of rewriting the application is a nice touch. makes this much easier to test on an existing project.
Inferock
@upendra_kumar17 that was a rule we set for ourselves, if trying it takes more than a minute nobody will ever find out what their calls really cost. Two settings in, and your app never knows the difference. Would be curious how it goes on your existing project.
Thanks @fmerian for hunting us. Feel free to ask if anybody has any questions about Inferock Bench.
Buffup.AI
The retry tracking caught my eye. silent retries are probably one of the easiest ways for API costs to creep up without anyone noticing.
@sansa_grey Right, retries are the sneakiest of the bunch: the answer still arrives, everything looks fine, and the bill just grows. Making those visible was one of the first things we wanted for ourselves.
I like that it works as a local proxy. Keeping billing and usage data on the developer's machine feels like a thoughtful design choice.
@awesome_america Thank you! That part was non negotiable for us, it's your spend and your keys, so the record should live on your machine, not on ours.
Tracking retries alongside failures is a smart addition. Those hidden retries can quietly become a big part of the bill.
@desire_waterman Thank you! That matches our experience, no single retry looks expensive, it's the accumulation that stings. Glad that part stood out to you.
What I really want to know is how this handles historical data. Can I feed it a month of past logs and get a retroactive receipt or is it strictly forward-looking from install? I ask because the overspending I'm most curious about already happened and I'd love a way to audit it after the fact.
Inferock
@kimberly_west really good question. Today it works by sitting in front of your live traffic, so the receipts start from the moment you point your SDK at it, it can't vouch for calls it never saw.
Bababot
I like that Inferock Bench focuses on evidence rather than simply showing another dashboard. Having a separate record of every API call could make unexpected billing much easier to investigate.
Inferock
@aarav_pittman thank you! Dashboards summarize, and summaries are where the weird stuff goes to hide. We wanted something closer to a paper trail, boring on purpose, there for the day you need to investigate.
Which API call actually cost me this money? is a question every AI app eventually needs to answer.
Inferock
@santosh__kumar9 exactly, every team hits that question sooner or later. We just want the answer already sitting there when it happens.
The two setting setup makes this feel unusally easy to try.
Inferock
@jordan_bulk thank you! Hopefully the receipts earn it the permanent spot.