Terms of Service
Effective date:
1. The service
shuntio provides OpenAI- and Anthropic-compatible access to large language models through a single endpoint. You call models served by upstream providers through us; we handle forwarding, metering and settlement.
shuntio is not a model provider. It does not train models, does not produce output of its own, and is not affiliated with any vendor whose models it routes.
By registering for or using the service you agree to these terms and to the Privacy Policy. If you do not agree, do not use the service.
2. Accounts
Accounts are created through Google or GitHub OAuth — signing in opens the account, with no approval step.
- You are responsible for every request made with a key issued to your account, including keys you create for other people.
- A key is shown once at creation and cannot be recovered in plaintext. Guard it like a password; never commit it to a repository or ship it inside a client application.
- If a key leaks, revoke it in the console — revocation is immediate — and tell us.
- One account per person or organisation, and accounts are not transferable.
3. Billing and credit
- Credit is prepaid and deducted per token, with input and output counted separately. Where a model supports prompt caching, cache reads and writes are billed on their own line.
- The pricing page and console reflect current rates. The rate applied to a call is the rate in effect at the moment of that call; a price change never reaches backwards.
- No minimum spend — no calls, no charges. An idle account costs nothing, and credit does not expire.
- Credit is a virtual metering unit usable only inside this service. It cannot be withdrawn, transferred, or treated as currency or a security.
- Once a payment succeeds and the credit arrives, refunds are generally not offered. Duplicate charges, or misstatements caused by a fault on our side, are refunded to the original payment method or re-credited after verification.
- A call that fails before reaching the upstream provider is not billed. A call the upstream provider already served is billed even if the answer was unhelpful, because the tokens were consumed. Raise a billing dispute by email within 60 days of the charge.
4. Acceptable use
You must not use the service to:
- break the law, or help anyone else do so;
- generate or distribute content that infringes intellectual property, privacy or other rights of others;
- attack, probe or reverse-engineer the service or its upstream providers, bypass metering or rate limits, or otherwise interfere with normal operation;
- sell or rent your account or keys, or resell API access without a written agreement with us;
- impersonate others or obtain credit through a false identity;
- process payment card data or other data you are not licensed to handle.
5. Availability and changes
There is no SLA. The service is provided on a best-effort basis. No uptime percentage, latency or error-rate target is promised, and no service credit is offered for downtime.
Upstream providers are third parties with their own outages, rate limits and policy changes. A failure upstream is passed through to you. If your workload needs a contractual availability guarantee, buy directly from the model vendor instead of from a gateway.
The model catalog and prices change as upstream providers adjust. Where a change withdraws a model you are actively using, we will say so in the console rather than removing it silently.
We may throttle, suspend or terminate accounts that abuse the service.
6. Third-party models and trademarks
Model names, model families and vendor names that appear on this site are the trademarks of their respective owners and are used only to describe what the service can call.
shuntio is an independent third party. It is not affiliated with, endorsed by, sponsored by, or authorised by Anthropic, OpenAI, Google, xAI or any other model vendor. Nothing on this site should be read as a claim of partnership.
The accuracy and appropriateness of model output is determined by the upstream model. Review generated content before relying on it, and take responsibility for decisions based on it.
7. Limitation of liability
The service is provided "as is" and "as available", without warranties of any kind, whether express or implied, including fitness for a particular purpose, accuracy of output, or non-infringement.
To the maximum extent permitted by law, we are not liable for indirect, incidental, special or consequential damages, or for lost profits, revenue or data. Our total liability arising out of or relating to any single dispute is capped at the amount you actually paid for the service in the 12 months preceding it.
8. Termination
You may stop using the service and request deletion of your account at any time.
Accounts that breach these terms may be suspended or terminated. Any remaining credit on a terminated account is not refunded, and a terminated user may not register again.
9. Changes to these terms
If these terms change, the effective date at the top of this page changes with it. Continuing to use the service after a change takes effect constitutes acceptance of the updated terms.
10. Contact us
Questions about these terms: support@shuntio.com.
This document is a plain-language baseline for a metered API relay service. It is not legal advice.