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
Hey PH
We built inferock-bench because we kept paying for AI answers that died mid sentence, and nobody could tell us where the money went.
Providers give you totals. They don't give you the per call receipt you'd need to prove which answer broke, which retry ran, or which token count changed. The company that charges you also decides what counts as a failure and keeps the only detailed records.
inferock-bench runs locally as a proxy in front of OpenAI, Anthropic, Gemini, OpenRouter shaped calls. Point your existing SDK at it (change two settings: apiKey and baseURL), and it captures every call as an independent, per call record. Your provider key never touches our servers, it's used locally only, attached to provider requests.
What it catches:
- Answers cut off mid stream that still got billed
- Empty replies with billed tokens attached
- Token counts that don't match visible output
- Retries that may have silently doubled a charge
- Cache discounts you may be missing on your invoice
Every run reports a receipt: spend observed, bill-bounded money loss, time loss, and a separate "invoice-check exposure" line that never gets summed into money loss, because we don't want a louder headline at the cost of a weaker claim.
Run it in about a minute: npx inferock-bench
Its open source (FSL-1.1-Apache-2.0, converts to Apache-2.0 in 2 years).
Question for this community: has anyone here actually disputed an AI provider bill and gotten a credit? What worked?
@bharath_koneti @tehreem_fatima5 Appreciate you actually answering the question, that matches what we kept hearing from others too. Without per request records the conversation with a provider ends pretty fast. You walk in with evidence instead of a feeling.
@tehreem_fatima5 Thank you!
Genuinely hoping it saves a few of those support conversations.
Congrats on the launch @himashwetha_gowda. Good find @fmerian.
Question regarding accuracy, is it possible it might not be able to distinguish a genuinely billable provider failure from valid hidden token usage, such as reasoning, refusal, cache, or tool-call tokens?
@rohanrecommends thank you! Anything that could be legit hidden usage like reasoning, cache, or tool tokens doesn't count as loss, it just gets flagged for you to check. Only what we can prove makes the loss number.
The failed-calls and retries breakdown is the part I'd use first. One case I keep hitting might not show up there: a tool call that returns 200 with a silently corrupted value. I measured this on Anthropic models, 40 calls, none flagged the value was wrong, so it bills as a clean success and the retry logic never fires. Can the receipt catch a call that looked fine but wasn't? Or is that out of scope by design?
@himashwetha_gowda Thanks, that's a clear answer and the right boundary to hold. The corrupted-200 case really does belong to a different layer. It needs a schema or an oracle on the payload and only the caller knows what the value should be. What you're proving from the outside is exactly the part I can't check myself today, so I'd keep it that sharp. Following the launch.
Love that this sits as a local proxy instead of asking me to route traffic through another cloud service. Keeps my API keys and data where they belong.
@peter_victor everything stays on your machine, we never see it.
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.
@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!
Thanks @fmerian for hunting us. Feel free to ask if anybody has any questions about Inferock Bench.
Does anyone know of any tools that can work with web-based logins (claud.ai etc)
@jay_janarthanan1 we don't touch that layer. inferock-bench works where there's an API key and a baseURL to point somewhere, and web app subscriptions don't expose the per call detail we'd need. If someone has cracked that curious to see it too.
auditing your own inference bill is such an obvious gap 🔍 nice one. seeing much overbilling in the wild yet?
Retries are the part I’d put in giant font. ‘$0.04/call’ means nothing if a successful job secretly takes 6 calls. I want cost per accepted outcome.
Congrats on the launch of Inferock Bench.
Love the idea of having an independent receipt for every LLM API call. Transparent usage, billing, and overpayment tracking can be a huge win for teams working with multiple LLM providers.