Terms of Service

OpenSourceAIHub.ai is a multi-modal AI gateway and security governance platform. The Service provides two primary operational modes:

Age: You must be at least 18 years of age.

Account: You are responsible for all activity under your API keys and account. You must notify us immediately of any unauthorized use.

Solopreneur Notice: You acknowledge that this Service is maintained by a limited team and “Uptime” is provided on a best-effort basis.

Heuristic Nature: Our DLP and PII redaction services (including OCR) use heuristic and machine-learning patterns. We do not guarantee 100% detection of sensitive data.

Best Effort: The Service is a “best-effort” security layer. It is not a substitute for proper data handling practices.

No Liability for Leaks: Datum Fuse LLC shall not be liable for any damages arising from the failure of the Service to detect or redact specific data points.

Output Accuracy: AI models are known to “hallucinate.” We are not responsible for the accuracy, legality, or safety of the outputs generated by the underlying LLMs.

Prepaid Credits: Credits purchased for the “Wallet” are not currency, have no cash value, and are non-transferable.

No Refunds: All sales of prepaid credits and Pro subscriptions are final and non-refundable, except as required by Illinois law.

Expiration: Credits will expire after twelve (12) months of account inactivity.

Pricing Changes: We reserve the right to change our retail token pricing at any time to reflect changes in upstream wholesale costs.

You agree NOT to use the Service to:

• Generate illegal content, child sexual abuse material (CSAM), or content that promotes terrorism.
• Attempt to reverse-engineer the Service or the underlying models.
• Exceed the rate limits set for your tier.
• Use the Service in a way that violates the Terms of Service of the underlying providers (including but not limited to OpenAI, Google, xAI, Groq, Together.ai, DeepInfra, Mistral AI, Amazon Bedrock, and other supported providers).

The Service: All software, code, “IP Guard” logic, and UI elements are the exclusive property of Datum Fuse LLC.

Your Data: We claim no ownership over the prompts you send or the completions you receive.

No Retention: As per our Privacy Policy, we do not persist or store the content of your prompts or completions once the inference request is fulfilled.

The Service relies on third-party providers for AI inference. We are not responsible for the availability, latency, or data-handling practices of third-party providers (including but not limited to OpenAI, Google, xAI, Groq, Together.ai, DeepInfra, Mistral AI, Amazon Bedrock, and other supported providers).

Force Majeure: We are not liable for service interruptions caused by upstream provider outages or cloud infrastructure failures (AWS).

You agree to defend, indemnify, and hold harmless Datum Fuse LLC and its employees, contractors, and officers from and against any and all claims, damages, obligations, losses, liabilities, costs, or debt, and expenses (including but not limited to attorney's fees) arising from your use of the Service or your violation of these Terms.

These Terms shall be governed by and construed in accordance with the laws of the State of Illinois, without regard to its conflict of law provisions. Any disputes shall be resolved exclusively in the state or federal courts located in Illinois.

We reserve the right to modify or discontinue the Service at any time. We may update these Terms from time to time; your continued use of the Service after such changes constitutes acceptance of the new Terms.

In Governance (BYOK) Mode, Customer may provide API credentials for third-party AI providers. Such credentials are stored in encrypted form and used solely for the purpose of processing requests initiated by the Customer through the Service.

The Company:

• Does not use Customer API credentials for any purpose other than automated request processing;
• Does not access Customer API credentials manually except as reasonably necessary for support or security purposes;
• Does not share Customer API credentials with any party other than the designated upstream provider required to process Customer requests.

Customer is solely responsible for safeguarding its API credentials and for all activity conducted using such credentials.

The Company reserves the right to suspend, restrict, or terminate access to the Service at any time, without prior notice, if it determines in its sole discretion that:

• Customer use violates these Terms;
• Customer use presents legal, regulatory, or security risk;
• Fraudulent, abusive, or excessive usage patterns are detected;
• Payment disputes, chargebacks, or non-payment occur;
• Platform stability or integrity may be impacted.

Suspension may include freezing Wallet balances during investigation of suspected fraud or abuse.

To ensure platform stability and prevent abuse, the Company may implement rate limits, throttling mechanisms, token caps, daily spend limits, or other technical controls.

Customer acknowledges that requests may be delayed, throttled, or rejected if usage exceeds permitted thresholds.

The Company shall not be liable for interruptions resulting from such enforcement mechanisms.

The Service may temporarily process prompt or request content in memory for the purpose of policy enforcement, routing, validation, and response generation.

Such processing does not constitute persistent storage, and prompt or completion content is not retained beyond the duration necessary to fulfill the request, except as otherwise disclosed in the Privacy Policy.

In the event of a payment dispute, chargeback, or suspected fraudulent transaction:

• The Company may suspend or terminate access to the Service;
• The Company may freeze Wallet balances pending resolution;
• The Company may recover disputed amounts from available Wallet credits.

Customer is responsible for resolving disputes directly with the payment processor.

Unless expressly agreed in writing, the Service does not include any service-level agreement (SLA) or uptime guarantee.

All availability is provided on a best-effort basis.

In BYOK Mode, Customer maintains a direct billing relationship with third-party AI providers.

The Company:

• Does not monitor Customer provider billing accounts;
• Is not responsible for charges incurred directly with such providers;
• Does not guarantee provider pricing accuracy.

Customer is responsible for reviewing and managing its third-party billing activity.

These Terms constitute the entire agreement between Customer and the Company and supersede all prior or contemporaneous communications and proposals.

If any provision of these Terms is found unenforceable, the remaining provisions shall remain in full force and effect.

Customer may not assign or transfer any rights under these Terms without prior written consent of the Company.

The Company may assign its rights and obligations in connection with a merger, acquisition, or sale of assets.

The Service utilizes a best-effort algorithmic router ("Smart Router") to select AI providers based on a combination of price, latency, and availability. Datum Fuse LLC does not guarantee that the Service will always select the absolute lowest-priced provider available on the global market at the time of a request. The User acknowledges that:

• Provider pricing is volatile and may change without notice. The Hub's internal pricing data is periodically refreshed and may have minor delays relative to real-time provider rate changes.
• The Smart Router optimizes for a balance of cost-efficiency, latency, and system reliability. In some cases, a slightly higher-cost provider may be selected to ensure acceptable response times or availability.
• Routing decisions are made using internal market data and algorithmic cost selection. The Company makes no representation that its pricing index is exhaustive of all global provider rates.
• No refunds will be issued based on price discrepancies between the Service and third-party providers.

All Managed Mode usage includes a transparent service fee: 25% markup for open-weight models (Llama, Mistral, Qwen) and 30% markup for closed-source models (OpenAI, Gemini, xAI). These markups cover security infrastructure, DLP enforcement, audit logging, and platform maintenance.

Customer acknowledges that usage-based services may result in rapid consumption of Wallet balances.

Customer is responsible for configuring usage limits and monitoring consumption.

The Company shall not be liable for depletion of Wallet balances resulting from Customer-initiated requests.

For questions regarding these Terms, please contact legal@opensourceaihub.ai.