Terms of Service
Last updated: 5 June 2026
1. Operator
These Terms of Service apply to the website, platform, compute credits, GPU capacity services and related support services operated under the brand HarborMesh Cloud.
Legal operator: DOUBLE J PARTNERS LIMITED. Company No.: 77498577. Registered office: RM 409, Beverley Commercial Centre, 87-105 Chatham Road South, Tsim Sha Tsui, Hong Kong.
2. Services
HarborMesh Cloud provides distributed GPU compute services, prepaid GPU compute usage allocation, capacity reservations and enterprise virtual datacenter arrangements for approved workloads such as AI inference, image and video generation, rendering, model serving, data processing and batch compute. Customers purchase direct access to cloud compute services from DOUBLE J PARTNERS LIMITED.
Services may be delivered through a combination of owned infrastructure, cloud infrastructure and verified third-party infrastructure suppliers. These suppliers act only as fulfillment providers, not as sellers, merchants, sub-merchants, payment recipients or settlement beneficiaries.
3. No regulated financial or crypto service
HarborMesh Cloud does not provide marketplace services, payment facilitation, merchant acquiring, payment processing for third parties, marketplace settlement, sub-merchant services, remittance, wallet services, e-wallets, e-money issuance, stored value facilities, custody, cryptocurrency exchange, cryptocurrency custody, investment products, mining-as-a-service or financial advice. Any prepaid usage allocation is non-cashable, non-transferable, non-redeemable and usable only for eligible GPU compute workloads.
4. Accounts and onboarding
We may require account registration, business verification, payment verification, usage screening, sanctions screening, workload review or additional onboarding before providing access to the services. We may reject or suspend an account if we consider the account, payment method, user, workload or requested activity to be unsafe, unlawful, unsupported or outside our commercial risk policy.
5. Acceptable use
You must not use the services for unlawful, harmful, infringing, abusive or high-risk activity. Prohibited activity includes malware, unauthorized access, credential theft, spam, phishing, evasion of security controls, illegal surveillance, unlawful scraping, generation of illegal content, infringement of intellectual property rights, regulated financial activity without approval, or workloads that violate applicable export control, sanctions, privacy, platform or payment provider rules.
6. Customer data and workloads
You are responsible for the legality, accuracy, security and rights clearance of any data, prompts, files, models, scripts, containers, outputs or workloads you submit. Unless we agree in a signed enterprise data processing agreement, you must not submit highly sensitive personal data, payment card data, medical records, state secrets, regulated financial records, confidential government data or other data requiring special handling.
7. Compute credits and billing
Compute credits are prepaid digital service credits used to access eligible GPU workloads. Credits are deducted according to workload type, GPU tier, runtime, queue priority, reservation level, failed job handling and any applicable order terms. Prices are displayed in Hong Kong dollars unless stated otherwise. Taxes, bank fees, payment fees and currency conversion charges may apply.
8. Enterprise capacity
Enterprise GPU capacity, dedicated virtual datacenter plans and high-volume daily plans may require a signed order form or statement of work. Availability, GPU tier, throughput, service levels, support response times, start date, minimum commitment, payment schedule and refund rules will be specified in the relevant order.
9. Service availability
Distributed GPU capacity may vary based on supply, provider availability, network conditions, workload compatibility and demand. We aim to provide commercially reasonable service continuity, but we do not guarantee uninterrupted access unless a separate signed SLA applies.
10. Outputs and workload results
We do not guarantee that outputs generated by AI models, rendering pipelines or third-party compute processes will be accurate, suitable, lawful, non-infringing or fit for a particular purpose. You are responsible for reviewing, validating and using outputs appropriately.
11. Approved infrastructure supplier program
Where services are delivered through verified third-party infrastructure suppliers, we may route workloads according to tier, performance, price, region, availability and risk controls. Infrastructure suppliers provide compute capacity under commercial supplier agreements and are not sellers, merchants, sub-merchants, payment recipients or settlement beneficiaries. We may benchmark, monitor, suspend or remove suppliers to protect service quality and customer trust.
12. Payment, failed payments and chargebacks
Payments may be processed by third-party payment service providers. You must provide accurate billing details and are responsible for any authorized purchases, invoices, taxes and fees. We may suspend access for failed payments, suspected fraud, payment disputes, chargebacks or compliance review.
13. Refunds
Refunds are governed by our Refund Policy. Used compute, completed workloads, started reserved capacity and custom enterprise work are generally non-refundable except as stated in the Refund Policy or a signed order form.
14. Intellectual property
We retain all rights in the platform, website, software, documentation, marks, systems, user interface, pricing logic, routing logic and service design. You retain ownership of your customer content, subject to the rights necessary for us to process, secure, transmit, store and display it for service delivery.
15. Confidentiality
Enterprise customers may request a mutual non-disclosure agreement before sharing sensitive commercial information. Public website use and self-serve accounts are not confidential unless separately agreed in writing.
16. Disclaimers
The services are provided on an “as is” and “as available” basis to the maximum extent permitted by law. We disclaim implied warranties of merchantability, fitness for a particular purpose, non-infringement, uninterrupted operation and error-free performance.
17. Limitation of liability
To the maximum extent permitted by law, we are not liable for indirect, incidental, special, consequential, punitive or exemplary damages, loss of profits, loss of data, loss of goodwill, business interruption or procurement of substitute services. Our aggregate liability for a claim is limited to the amount paid by you for the affected services in the three months before the event giving rise to the claim, unless a signed enterprise agreement states otherwise.
18. Suspension and termination
We may suspend or terminate access if we believe you violated these Terms, created operational or legal risk, submitted unsupported workloads, failed to pay, initiated improper chargebacks or used the services in a way that could harm us, other users, providers, payment partners or third parties.
19. Governing law
These Terms are governed by the laws of the Hong Kong Special Administrative Region. Unless otherwise agreed in a signed enterprise agreement, disputes will be subject to the non-exclusive jurisdiction of the courts of Hong Kong.
20. Contact
For legal notices, contact legal@harbormeshcloud.com. For support, contact support@harbormeshcloud.com.