Questions about accountable inference

Sign up to create your Inferock account, then choose how you run. With Inferock Watch, you connect your own provider keys. With Inferock Serve, you prepay through Inferock, and models and service terms are confirmed during onboarding. Inferock Dedicated is configured and priced by agreement before provisioning. Creating an account does not by itself make a model, region or capacity available.

Inferock is reliable and accountable LLM inference. You route selected model calls through Inferock; it records independent per-call receipts and, in Inferock Serve mode, can apply bounded service credits for eligible objective provider failures under agreed terms.

Inferock Dedicated is dedicated inference: capacity dedicated to your workload, with the configuration, price, commitment and service level agreed for your deployment before capacity is provisioned. Inferock bills you under your agreed capacity quote. To start, create your account, then discuss your workload with us.

Inferock Watch is FREE on your own provider keys. Inferock Serve has a 10% total service fee on actual supplier cost for every account. Inferock Dedicated pricing is By agreement.

Inferock records the request path, provider response, timing, usage, charge context, and failure evidence. The report shows what failed, what it cost, and whether the row is report-only or eligible for Inferock Serve service credit. Support.

Signup credit, if recorded for your account, is a one-time gift separate from credit earned on eligible Inferock Watch failures. No card, provider connection, paid tier or minimum spend is required to receive it. Signing in again or switching products does not add another signup credit.

Check Credits for the recorded amount, remaining balance, amounts applied or expired, and the entry’s expiry date. Signup credit is non-cash and expires. It can apply to eligible supplier costs on Inferock Serve; the service charge stays cash.

Downtime, provider timeouts, empty output, truncated output, invalid structured output or tool call arguments when a schema exists, and material billing or cache anomalies with observable evidence.

Credit requires verified, billed, unrecovered failures. Refusals, unbilled calls and fully recovered costs earn no credit. The wallet cap is 1,000 USD and up to 1,000 USD may be applied per calendar month; the service charge stays cash.

There is no blanket end-to-end encryption guarantee. Our intended confidential route is designed to encrypt content end to end between your client and a verified confidential workload. In that design, the model processes decrypted content inside the workload. Each route still requires qualification evidence for its exact model and configuration. This intended protection does not apply to ordinary Inferock Watch connections.

Provider credentials are encrypted for storage and tied to your account. Plaintext provider keys are not logged or returned after activation.

Run inferock-bench locally, send metered provider API traffic through it, and review the receipt. The receipt leads with spent dollars, bill-bounded money loss, time loss, and invoice-check exposure, then keeps provider-recognized recovery, recognition gap, and invoice-check exposure detail separate. It preserves token usage, cost, failure, retry, and billing-integrity evidence; it does not replace provider invoice records. Receipts + ledger.

Start by separating usage volume, token mix, retries, failed calls, cache behavior, and pricing. `inferock-bench` can record metered traffic that passes through the local proxy and show per-call token usage, cost, failures, retries, cache signals, the spent/money-loss/time-loss/invoice-check-exposure headline, and the invoice-check exposure detail. It cannot explain provider charges for traffic it did not see, and it does not replace the provider invoice. Compare the documented pricing rungs.

Only for traffic it observed. If calls passed through `inferock-bench`, the local receipt can show the measured request, response, usage, timing, retry, and cost evidence. If a provider key was used outside the proxy, the bench cannot reconstruct that unobserved traffic. Revoke exposed keys, preserve provider dashboard and invoice records, and use `inferock-bench` to measure future metered calls. See where observed evidence appears.

Inferock records the provider response, usage fields, timing, delivered content, error status, and price evidence for the call. When charge evidence is present, it can show bill-bounded money loss for a failed, incomplete, timed-out, or 504 call. When the evidence is only an invoice-check signal, such as cache discount at risk, the receipt labels it as invoice-check exposure instead of money loss. If the provider exposes too little charge evidence, the receipt labels that limit instead of inventing a refund claim. Check the documented credit conditions. Support.

Inferock can preserve local evidence for OpenAI token, usage, retry, and failure questions. It reports evidence strength, separates bill-bounded money loss from provider-recognized recovery and recognition gap, and keeps invoice-check exposure out of the money-loss total instead of declaring every mismatch an overcharge.

Inferock can record Anthropic usage fields, delivered output, pricing evidence, and token cross-check signals where available. The result is a receipt-grade evidence trail, not an automatic accusation.

Use inferock-bench as a localhost proxy and point your SDK base URL at it. Provider keys stay local, and the bench records token usage and cost evidence for supported OpenAI, Anthropic, Gemini, and pinned OpenRouter calls. Follow the first-call guide.

Yes. inferock-bench is a local LLM cost-tracking proxy for development and evaluation traffic. It turns provider calls into per-call receipts with spent dollars, bill-bounded money loss, time loss, invoice-check exposure, failure evidence, and billing-integrity signals.

It is a measurement proxy, not a provider-wide spending firewall. For traffic routed through the local bench, it records retries, duplicate request evidence, token usage, timing, and cost so a retry loop or agent run leaves a receipt. The built-in test flow shows an estimate and spend cap before it starts provider calls. Your provider account still needs its own key controls, quotas, and billing limits. Review the gateway limit behavior.

The public benchmark measures observable reliability and billing signals on real provider traffic. It is not a provider ranking unless a run explicitly reports comparable measured data and coverage.

It records timing for calls routed through the bench and can report latency against configured thresholds. It can show that a call was slow, timed out, or failed from the client-visible evidence. It cannot see provider internal queues or prove an internal cause unless the provider exposes that evidence. Inspect the recorded timing evidence.

It can record the requested model, exposed model/version fields, output shape, refusal signals, latency, and configured drift probes. A quality change becomes useful evidence only when there is an approved baseline, threshold, or ground truth. Broad complaints like "the model got worse" stay report-only until the receipt has that baseline.

Not as a general in path failure class. Factuality needs ground truth: a reference answer, retrieval corpus, database result, human label, or downstream verifier. Without that, Inferock can only flag weaker signals.

Failure type, event count, estimated cost, evidence, and remedy status. Open the sanitized sample scorecard.