Terms of Service

By creating an account, signing in, buying a plan or balance, creating an API key, or otherwise using TunAI (the “Service”), you agree to these Terms, to the Privacy Policy, and to the Cancellation & Refund Policy and Service Delivery Policy, which are part of these Terms. If you do not agree, do not use the Service.

You must be at least 18 years old, or at least 13 and using the Service with the consent and under the supervision of a parent or legal guardian who accepts these Terms on your behalf. Paid plans, top-ups and the developer API are available only to adults. If you use the Service on behalf of a company or another organisation, you confirm that you are authorised to bind it, and “you” means that organisation.

Before you can create an API key we ask you to confirm, in the developer console, that you have read and accept the version of these Terms then in force. We record the version and the time of that acceptance on your account.

TunAI is a gateway. It gives you one account, billed in Tunisian dinar, through which you can use large language models operated by third-party providers (OpenAI, Anthropic, Google, Mistral AI, Groq, DeepSeek, Zhipu AI and Alibaba Cloud, among others) in two ways:

• The chat app at tunai.tn and in our iOS and Android apps: conversations, file and image uploads, web search, sharing.
• The developer platform: an OpenAI-compatible HTTP API at https://api.tunai.tn/v1 and a console at https://console.tunai.tn where you create keys, watch usage and top up.

We do not build or train the models. We route your request to the provider of the model you chose, meter the result and return it to you. Which models are available, what they cost and how they behave can change when the providers change them; we may add, replace or retire models at any time and will not be liable because a particular model is slower, different or no longer offered.

• Sign-in is provided by Auth0. Keep your sign-in method secure; everything done through your account is treated as done by you.
• Give us accurate information and keep it current. We may ask you to verify your identity or your payment method, and may suspend an account we cannot verify.
• One person, one account. Do not share, sell or transfer an account, and do not create accounts automatically or to get around a limit, a suspension or a ban.
• Tell us at support@tunai.tn as soon as you suspect unauthorised use of your account or a key.

These additional terms apply when you use API keys, the console or the API.

• Keys are secrets. A key gives whoever holds it access to your account’s balance. Keep keys out of client-side code, public repositories and shared documents; use per-project keys with scopes, expiry dates and monthly budgets. You are responsible for every request made with your keys, whether or not you authorised it, until you revoke them in the console.
• Limits. Each key is limited to a number of requests per minute (the default is shown in the console) and to whatever budget, scope and expiry you set. We may apply further limits to protect the Service and may throttle or reject traffic that looks automated abuse.
• Compatibility. The API speaks the OpenAI chat-completions, embeddings and models format. Other formats (including the Anthropic Messages API) are not supported unless the documentation says so. We aim to keep the API stable but may change or retire endpoints, parameters and models; material changes will be announced in the console or by email where practical.
• Your applications. If you build a product on the API, you are the operator of that product: you must have your own terms and privacy notice, comply with the law that applies to your users, not present model output as coming from a human where that would mislead, and make sure your users cannot use your product to do anything the Acceptable Use Policy forbids. You may not resell raw access to the API, or offer a service whose main purpose is to proxy it, without our written agreement.
• Moderation is your responsibility. We do not screen API requests or responses. You alone are responsible for everything sent through your keys and everything your application does with the output, including moderating your own users’ input and the output you show them, and for the consequences of failing to do so. Section 6.1 applies in full. A provider may still refuse a request under its own policy.
• Logging. For each API request we store the model, token counts, cost, timing, outcome, request id and any error message. We do not store the content of API requests or responses. See the Privacy Policy.
• Provider policies. Every model is operated by a third party whose usage policy binds us and, through these Terms, you. You agree to comply with the policy of each provider whose models you use, including: OpenAI, Anthropic, Google, Mistral AI, Groq, DeepSeek, Zhipu AI (Z.ai), Alibaba Cloud (Qwen). Where a provider’s policy is stricter than ours, the stricter rule applies to requests sent to that provider.

You may not use the Service, directly or through an application you build on it, to:

• break any law that applies to you or to us, or help someone else do so;
• create, request or distribute sexual content involving minors, or any content that sexualises children — we report this to the authorities;
• promote, plan or facilitate violence, terrorism, human trafficking, or serious harm to people or property;
• develop or spread malware, or attack, probe or overload any system, including ours and the providers’;
• defraud, scam, phish, impersonate a person or organisation, or generate spam, fake reviews or deceptive political or commercial content;
• harass, threaten, bully or defame anyone, or generate hateful content that targets people on the basis of who they are;
• collect, infer or publish personal data about people without a lawful basis, including doxxing, covert profiling and facial or biometric identification;
• infringe copyright, trademarks, trade secrets or other rights, or circumvent technical protection measures;
• make automated decisions with legal or similarly significant effects on people (credit, employment, housing, insurance, healthcare, education, law enforcement, immigration) without meaningful human review, or provide medical, legal or financial advice that presents itself as coming from a licensed professional;
• build weapons, or provide instructions for chemical, biological, radiological or nuclear harm;
• bypass or probe safety systems, rate limits, billing or usage limits, share accounts or keys to avoid limits, or extract data from the Service by scraping or reverse engineering;
• use the Service from, or for the benefit of, a person, entity or territory subject to sanctions or export controls that the model providers are required to observe.

We may investigate suspected violations, remove content, throttle or suspend accounts and keys, and report conduct to the authorities where the law requires or allows it. Chat messages are screened automatically (section 6.2). Beyond that, no person reads your conversations to police this policy; we act on reports, on automated signals such as moderation results, provider refusals and abuse patterns, and on legal requests.

Large language models generate text statistically. Their output can be inaccurate, incomplete, out of date, biased, offensive or simply invented, and it can differ from one run to the next. It is not professional advice of any kind — medical, legal, financial, tax, engineering or otherwise — and it is not a statement by us. You are responsible for reviewing output before you rely on it, publish it, or show it to your own users, and for any decision you make on the basis of it. Do not use the Service as the sole basis for decisions that could hurt someone or that the law reserves to a qualified person.

We do not control how the providers’ models behave, and we do not guarantee that any output will be free of third-party rights or usable for a particular purpose.

The Service depends on providers we do not control: the model providers, Auth0 for sign-in, ClicToPay for payments, and hosting and monitoring providers listed in the Privacy Policy. Their availability, changes and terms affect the Service. We are not responsible for their acts, omissions or content, and any link to a third-party site is provided for convenience only.

We provide the Service on a best-effort basis, without a contractual uptime commitment. We may modify, suspend or discontinue any feature, model or plan, and may interrupt the Service for maintenance or for safety reasons. Where a change removes something you have paid for and have not yet used, we will give you the unused portion back as balance or as a refund, at our choice. Features marked beta, preview or experimental may change or disappear without notice.

By you. You may stop using the Service at any time and ask us to delete your account (see the Privacy Policy). Prepaid periods and balance are not refunded on deletion except as the Cancellation & Refund Policy provides.

By us. We may suspend or terminate your access, revoke keys and remove content, with notice where reasonably possible, if you breach these Terms or the Acceptable Use Policy, if your payment is reversed or fraudulent, if a provider or authority requires it, if the account creates legal or security risk for us or others, or if it has been inactive for more than 24 months. We may also end the Service or a plan altogether with at least 30 days’ notice, in which case we refund unused prepaid amounts.

On termination your right to use the Service ends, keys stop working and we delete your data as described in the Privacy Policy. Sections 5 to 8 and 12 to 16 survive termination.

THE SERVICE AND ALL OUTPUT ARE PROVIDED “AS IS” AND “AS AVAILABLE”. TO THE FULLEST EXTENT THE LAW ALLOWS, WE DISCLAIM ALL WARRANTIES, EXPRESS OR IMPLIED, INCLUDING MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE, ACCURACY, NON-INFRINGEMENT AND UNINTERRUPTED OR ERROR-FREE OPERATION. Nothing in this section limits the rights the law gives you as a consumer that cannot be waived.

TO THE FULLEST EXTENT THE LAW ALLOWS, AI GATEWAY SUARL, ITS MANAGER, EMPLOYEES, CONTRACTORS AND SUPPLIERS WILL NOT BE LIABLE FOR ANY INDIRECT, INCIDENTAL, SPECIAL, CONSEQUENTIAL OR PUNITIVE DAMAGES, OR FOR LOST PROFITS, REVENUE, DATA, GOODWILL OR BUSINESS OPPORTUNITY, ARISING FROM OR RELATED TO THE SERVICE OR ITS OUTPUT, HOWEVER CAUSED AND UNDER ANY THEORY OF LIABILITY, EVEN IF WE WERE TOLD SUCH DAMAGES WERE POSSIBLE.

OUR TOTAL LIABILITY FOR ALL CLAIMS RELATING TO THE SERVICE IN ANY 12-MONTH PERIOD IS LIMITED TO THE GREATER OF THE AMOUNTS YOU PAID US IN THE 12 MONTHS BEFORE THE EVENT GIVING RISE TO THE CLAIM AND 100 TND.

These limits do not apply to liability that cannot be limited under Tunisian law, including liability for fraud, gross negligence, or death or personal injury caused by our negligence, and they do not limit your non-waivable rights as a consumer.

You will defend, indemnify and hold harmless AI GATEWAY SUARL and its manager, employees and suppliers from any claim, loss, liability, fine and reasonable legal cost arising from your input, your output as you use it, an application you build on the API, your users, or your breach of these Terms, the Acceptable Use Policy, a provider policy or the law. This section does not apply to a consumer to the extent the law does not allow it.

These Terms and any dispute relating to the Service are governed by the laws of the Republic of Tunisia. If you are a consumer, you also keep the protections of the mandatory consumer law of the country where you live.

Before going to court, please write to legal@tunai.tn; we will try to resolve the matter with you within 30 days. Failing that, the courts of Tunis, Tunisia have jurisdiction, without prejudice to any mandatory rule that gives a consumer the right to sue or be sued elsewhere. Consumers in Tunisia may also refer a complaint to the competent consumer protection authorities.

We may update these Terms. The version and effective date appear at the top of this page. For material changes we will notify you by email or in the Service at least 15 days before they take effect, except where a change is required by law or by a provider and must apply sooner. Using the Service after a change takes effect means you accept it; creating a new API key always requires accepting the version then in force. If you do not agree to a change, stop using the Service and, if you have unused prepaid time or balance affected by the change, contact us.

• Entire agreement. These Terms, with the policies they incorporate, are the whole agreement between you and us about the Service and replace any earlier terms.
• Severability. If a provision is unenforceable, the rest stays in force and the provision is applied to the extent the law allows.
• No waiver. If we do not enforce a provision, we may still enforce it later.
• Assignment. You may not transfer these Terms or your account. We may transfer them to a successor of the business, with notice to you.
• Force majeure. Neither of us is liable for a failure caused by events beyond reasonable control, including outages at the providers, network or power failure, or acts of authorities.
• Notices. We send notices to the email on your account; you send notices to legal@tunai.tn.
• Language. These Terms are written in English. Translations we may publish are for convenience; the English text prevails.

AI GATEWAY SUARL, 39 Rue du Japon, Apt 1-2, 1st floor, Montplaisir, 1073 Tunis, Tunisia. Legal: legal@tunai.tn · Support: support@tunai.tn.