Terms of Service

Welcome to the distributed LLM training platform, GPU/CPU cloud servers, enterprise token distribution platform, GEO content optimization, and related integration, operations, consulting, and technical support services provided by NeuraSea (hereinafter "the platform" or "we").

By using this platform, users acknowledge that they have read, understood, and agreed to these terms. Where these terms conflict with separately signed contracts, orders, or supplementary agreements between the parties, the separately agreed written terms shall prevail.

1. Service Content

Services provided by the platform include but are not limited to:

• Distributed scheduling for LLM training, elastic GPU resources, and high-throughput data pipelines;
• GPU and CPU cloud servers for training / inference / business workloads, available on-demand or reserved;
• Multi-model aggregation access for enterprise AI, token distribution, quota and metering, billing and reconciliation;
• GEO content structuring for the generative AI era, AI search engine compatibility, and citation monitoring;
• Private deployment, dedicated nodes, custom integration, technical support, and consulting services.

The specific service scope, feature activation, SLA, delivery method, and pricing standards are subject to the platform's actual features, console configuration, orders, contracts, or documents confirmed in writing by both parties.

2. Account and Credential Management

Users should properly safeguard their accounts, passwords, API keys, access tokens, node credentials, and channel keys, and are responsible for operations, invocations, tasks, and charges initiated through their accounts or credentials.

Losses caused by improper safekeeping, credential sharing, improper authorization, internal management negligence, or unauthorized third-party use shall be borne by the user. Upon discovering credential leakage, misuse, or abnormal invocation risks, users should promptly disable, reset, or adjust permissions and contact us for assistance.

3. Usage Guidelines

Users shall not use this platform to engage in the following activities:

• Activities that violate laws, regulations, regulatory requirements, industry standards, or public order;
• Activities that infringe on others' intellectual property, trade secrets, personal information rights, or other legitimate rights;
• Generating or distributing false information, fraudulent content, malicious code, illegal or infringing content;
• Circumventing, damaging, or interfering with the platform's authentication, billing, rate limiting, auditing, and security mechanisms;
• Unauthorized access, scanning, attacking, or interfering with the platform, third-party model services, or other systems;
• Using platform resources for mining, distributed attacks, providing unauthorized forwarding services, etc.;
• Other activities that may harm the platform, third-party service providers, other users, or public interests.

The platform reserves the right to take measures including reminders, restrictions, suspension, or termination of service for abnormal usage, violations, or high-risk behavior based on security, compliance, risk control, or operational needs.

4. Third-Party Models and External Services

Users may access models and external capabilities provided by mainstream model vendors and cloud service providers through the token platform or self-built pipelines. The availability, pricing, output quality, rate limits, content policies, data processing methods, and service rules of third-party services are determined by the respective third parties.

Users should independently confirm the legality, applicability, and compliance requirements of their use of third-party models and external services, and comply with the respective third party's terms of service, privacy policies, content policies, regional restrictions, and billing rules.

5. Data and Content Responsibility

Users are responsible for prompts, datasets, training samples, context, files, API requests, model outputs, and business configurations submitted through the platform, and should ensure they have the legal rights or authorization to process, upload, invoke, forward, and use such data.

The platform processes relevant data only to the extent necessary for identity verification, request forwarding, model and training task execution, quota deduction, usage statistics, log auditing, troubleshooting, security protection, and service operations. Unless required by law, authorized by users, agreed in contracts, or necessary for service delivery, the platform will not use user business data for unrelated purposes.

6. Fees and Settlement

The platform may charge based on factors including resource type, node specifications, invocation volume, model type and multiplier, deployment method, service packages, and technical support scope. Specific pricing, billing cycles, complimentary quotas, overage rules, refund rules, invoice arrangements, and payment methods are subject to console display, orders, contracts, or documents confirmed in writing by both parties.

Users should pay relevant fees in a timely manner as agreed. If users have overdue payments, insufficient quotas, or abnormal settlement risks, the platform reserves the right to restrict or suspend related services.

7. Service Changes, Suspension, and Termination

To ensure service security, stability, compliance, and continuous optimization, the platform may adjust features, interfaces, models, nodes, scheduling strategies, billing rules, rate limiting rules, or service policies.

If users violate these terms, contractual agreements, laws and regulations, or third-party service rules, or engage in activities that endanger system security, default on payments, abuse resources, or infringe on others' rights, the platform reserves the right to take measures including restricting invocations, suspending accounts, closing related features, or terminating services, and reserves the right to pursue legal liability.

8. Disclaimer

AI model outputs are probabilistic and uncertain, and may contain errors, omissions, biases, incompleteness, or may not be suitable for specific business scenarios. The platform does not guarantee the accuracy, completeness, legality, reliability, or fitness for a particular purpose of model outputs.

Users should conduct necessary manual review, testing verification, and risk assessment based on business scenarios. For high-risk scenarios such as healthcare, finance, law, government affairs, industrial safety, and life and health, model outputs shall not be used as the sole basis for decision-making.

9. Intellectual Property

Software, interfaces, documentation, technical solutions, trademarks, logos, service names, and other intellectual property related to the platform, unless otherwise stated, belong to NeuraSea or relevant rights holders. Without authorization, users may not copy, modify, distribute, sell, rent, reverse engineer, or otherwise improperly use them.

Users retain their legally owned rights to business data, input content, and output content, but should ensure that such content does not infringe on third-party legitimate rights.

10. Contact Information

For inquiries about business cooperation, contract signing, invoices, private deployment, technical support, or terms of service matters, please contact us via business email at contact@neurasea.com.