Service Level Agreement

1. Introduction

This Service Level Agreement ("SLA") is made between GMI Cloud US Inc. ("Provider" or "GMI") and Utopai Studios Inc.[GL1] ("Client") for the provision of dedicated infrastructure services. This agreement outlines the services provided, performance expectations, and the responsibilities of both parties.

2. Scope of Services

• AI Infrastructure Provision
• Support services
• Regular Updates
• Professional Services

3. Support Services

The Provider provides support for the dedicated, high-performance AI infrastructure provided to the Client.

3.1 Support Channels

• Client tickets can be raised via the GMI Cloud portal on the GMI Cloud website or via email.
• A dedicated Slack channel [GL2] will be available for communication. Please note work will only be carried out via an associated Client ticket being raised.
• A dedicated phone number is provided for P1 (Critical issues) only.

3.2 Regular Updates

The Provider provides periodic system updates, patches and performance optimisations as standard for supplied services, notice for which will be provided in advance, as per details in Section 12.

3.3 Service Levels

Availability

• 24/7 Remote Support
• On-site support during business hours (08:00 – 19:00 CET)

Response Time

• Target response varies between 1 hour – 1 business day depending on severity
• Details are available in section 11.2

Support Channels

• All channels listed above (Note: all work carried out must have an associated Client ticket)

4. Professional Services

4.1 Initial Professional Services

• Duration: Up to 10 working days provided to all Clients after the delivery of the AI infrastructure platform.
• Working Hours: Services are available during business hours (08:00 – 19:00 CET).
• Additional professional services can be purchased and scheduled via GMI Support Team at a rate of $4000 USD [GL3] per day.

4.2 Eligible Activities

Professional services can be utilized for:

• Implementation Assistance: Helping to deploy and configure AI workloads.
• Optimization Services: Performance tuning and optimization of infrastructure.
• Technical Consulting: Advisory services on best practices and strategies.
• Training and Knowledge Transfer: Educating Customer teams on system use and maintenance.
• Project Support: Assistance with specific projects or tasks defined by the Client.

4.3 Limitations

• Availability: Professional services are limited to business hours.
• Unused Services: Any unused initial professional services hours expire if not utilized within 30 days post-delivery.
• Cancellation Policy: Cancellations or rescheduling require at least 48 hours' notice.

5. Service Level Objectives

5.1 Network Availability

The Provider endeavors to maintain a network availability of 99.5% in any given calendar month.

The Monthly Network Uptime Percentage is considered available whereby Provider's border routers and core-switches of the relevant data center, are responding to Provider's monitoring tools. The relevant monitoring tools check border routers and core-switches every 15 seconds, excluding any scheduled maintenance windows as measured by the Provider. The Service is unavailable if it has no external connectivity using the network access method provided to access the Service from the internet.

The Parties acknowledge that, due to limited resiliency at the Boden Data Centre, the current average uptime is approximately 99.5%. In the event of a power outage, PROVIDER shall use commercially reasonable efforts to restore operations at Boden Data Centre as soon as reasonably practicable.

If a power outage extends beyond five (5) consecutive hours, Provider may, at its sole discretion and subject to technical feasibility, evaluate potential options to migrate the Services to alternative clusters at other GMI-operated facilities. The timing, scope, and method of any such migration shall be determined exclusively by Provider, and Provider makes no warranty that migration will be possible in all circumstances.

Failure to be able to log in to the platform, or send data to the Service due to any of the following, is excluded from the Monthly Percentage Uptime calculation under this SLA:

• Downtimes caused by events outside of Provider's control, including but not limited to any force majeure events, failures, acts or omissions of our upstream providers or failures of the Internet.
• Downtime caused by a loss of power, based on the limited resiliency mentioned above.
• Unavailability that resulted from Client's software or hardware, or third-party software or hardware, or both.
• Client's failure to use required configurations or follow acceptable use policies.
• Client's use of the Service in a manner inconsistent with Provider documentation.
• Downtime experienced in the context of scheduled maintenance which Provider has announced as per section 12 of this agreement.
• Downtime caused by actions of third parties, including but not limited to security compromises, denial of service attacks and viruses provided Provider makes reasonable efforts to keep its software and systems up to date.

Downtime is calculated as the time from Provider receiving a defect report from the Client via the aforementioned support channels, to the network being confirmed as operational and accessible via the Provider's platform. Where on-site support is required outside of business hours, availability hours will only be counted as per business hours (08:00 – 19:00 CET).

5.2 GPU Availability

The Provider endeavors to maintain a GPU availability of 95% in any given calendar month.

Monthly GPU Availability is measured in minutes per calendar month and is considered available when the GPUs assigned to the Client are operational and accessible via Provider's platform, excluding any scheduled maintenance windows as measured by Provider (details on notice periods can be found in section 3.2).

The following are excluded from the Monthly GPU Availability calculation under this SLA:

• Downtimes caused by events outside of Provider's control, including but not limited to any force majeure events, failures, acts or omissions of our upstream providers or failures of the Internet.
• Downtime caused by a loss of power, based on the limited resiliency mentioned above.
• Unavailability that resulted from Client's software or hardware, or third-party software or hardware, or both.
• Client's failure to use any required configurations, or follow acceptable use policies.
• Client's use of the Service in a manner inconsistent with the Provider's documentation.
• Downtime experienced in the context of scheduled maintenance which Provider has announced as per section 12 of this agreement.
• Downtime caused by actions of third parties, including but not limited to security compromises, denial of service attacks and viruses provided Provider makes reasonable efforts to keep its software and systems up to date.

Downtime is calculated as the time from Provider receiving a defect report from the Client, via the aforementioned support channels, to assigned GPU's being confirmed as operational and accessible via the Provider's platform. The calculation will be limited to the GPUs impacted. Where on-site support is required outside of business hours, availability hours will only be counted as per business hours (08:00 – 19:00 CET).

6. Service Credits

6.1 Service Credit

Service Credit means credits given to Client based on Downtime, as calculated under clause 6.2 and 6.3 below.

6.2 Network Availability Service Credits

If the Provider fails to meet Network Availability uptime of 99% in any calendar month, the Client will be entitled to service credits as follows:

• Availability between 95.0% and 98.9%: Credit of 10% of the monthly fee for the affected service.
• Availability between 90.0% and 95.0%: Credit of 20% of the monthly fee for the affected service.
• Availability below 90.0%: Credit of 30% of the monthly fee for the affected service.

6.3 GPU Availability Service Credits

If the provider fails to meet the GPU Availability uptime of 95% in any calendar month, the Client will be entitled to service credits as follows:

• Availability between 90.0% and 94.99%: Credit 5% of the monthly fee for affected service.
• Availability between 80.0% and 89.99%: Credit 10% of the monthly fee for affected service.
• Availability below 80.0%: Credit 20% of the monthly fee for affected service.

7. Roles and Responsibilities

7.1 Provider Responsibilities

• Deliver services as per the agreed standards.
• Provide support according to the selected support level.
• Ensure compliance with industry best practices and security standards.
• Maintain clear communication channels with the Client.
• Provide initial and additional professional services as agreed.

7.2 Client Responsibilities

• Use the services in accordance with the Acceptable Use Policy.
• Provide necessary access and information for service delivery.
• Report incidents and issues promptly and via the agreed channels noted in section 3.3.
• Schedule and utilize professional services within the agreed periods.
• Adhere to cancellation policies for scheduled professional services.
• Not raise cases at a priority that is higher than the actual impact assessment.
• Co-operate fully with the Provider if more information is needed to ensure that it is provided 'on time', to mitigate the risk of breaching the SLA.

8. Industry Best Practice

Services will be delivered in adherence to industry best practices, including:

• ITIL Framework: For effective IT service management.
• ISO/IEC 27001: Ensuring robust information security management systems.
• Continuous Improvement: Regularly enhancing services based on feedback and performance metrics.
• Change Management Procedures: Managing updates and changes with minimal disruption.

9. Monitoring and Reporting

9.1 Network Monitoring

Continuous real-time monitoring of network performance and availability is in place as per definition in section 5.1.

10. Break/Fix Hardware Replacement

10.1 Coverage

Components Covered: The Provider will perform break/fix hardware replacements for all components that meet RMA criteria, including GPU nodes, networking equipment, and other associated hardware.

10.2 Break/Fix Response Times

Initial Response: Where a hardware failure is identified and the RMA criteria met, an RMA request will be made within 1 business hour.

Replacement Commencement: Any hardware replacement will be attempted within 1 business day; however, this is on a best endeavors basis.

10.3 Conditions

Hardware Availability: Replacement times are subject to hardware availability and may be affected by supply chain constraints.

11. Incident Management

11.1 Elements Covered

• Incident Reporting: Defined channels for reporting issues.
• Escalation Paths: Clear procedures for escalating unresolved incidents.
• Root Cause Analysis: Investigations conducted for critical incidents to prevent recurrence.
• Communication: Regular updates provided during incident resolution.

11.2 Incident Levels

Provider will provide a response to the ticket within the initial response time as set out in the table below.

Service Requests are requests for action or tasks that are not related to an incident. The Provider will target initial response times to service requests within 1 business day.

11.3 Incident Reports

Following any P1: Critical incident, a Post Incident Report will be made available within 5 business days.

12. Change Management

• There will be an allowable period of one hour of scheduled maintenance per month that may have an impact to service, which will be communicated within the notice periods below.
• There may be occasions when a longer period may be required. This will be communicated and arranged in advance with the Client.
• Prior to any change, a full impact assessment will be completed to evaluate any potential effects on services.

13. Service Delivery Support

13.1 Support level

There will be a dedicated Service Delivery Manager (SDM) 09:00 – 18:00 CET Mon – Fri, who will serve as the primary point of contact for all matters relating to service performance, governance, and continuous improvement. The SDM will be responsible for overseeing the effective delivery of services in alignment with agreed service levels and business objectives. As part of this engagement, the SDM will provide upon request a formal monthly service performance report. These interactions are designed to ensure proactive service management, timely issue resolution, and sustained alignment between the service provision and the Client's evolving needs.

14. Additional Terms

Subject to this SLA and the applicable Order, by providing Customer Data to Company, Client grants Company a license to host, store, transfer, display, perform, reproduce, modify for the purpose of formatting for display, and distribute Customer Data solely and exclusively for the purpose of providing the Services to Client. For the purposes of this Agreement, "Customer Data" means all data uploaded by Client to Company's infrastructure and all data created by Client to Company's infrastructure.

Client is solely responsible for its Customer Data and agrees that Company is not and will not in any way be liable for Customer Data. By providing Customer Data, Client affirms, represents and warrants that: (i) its Customer Data and use thereof will not violate this SLA, the applicable Order or any applicable law, regulation, rule or third party rights; (ii) Client is solely responsible for the development, moderation, operation, maintenance, support and use of Customer Data, including when Customer Data is provided by Client's end users; (iii) Client's Customer Data and its use thereof does not and will not: (1) infringe, violate, or misappropriate any third party right, including any copyright, trademark, patent, trade secret, moral right, privacy right, right of publicity, or any other intellectual property or proprietary right; (2) slander, defame, libel, or invade a right of privacy, publicity or other property rights of any other person; or (3) cause us to violate any law, regulation, rule, or rights of third parties; and (iv) except for the specific Services provided under this SLA and the applicable Order or other express contract, Client is solely responsible for the technical operation of Customer Data, including on behalf of Client's end users. "Customer Data" means all data uploaded by Client to Company's infrastructure and all data created by Client to Company's infrastructure.

In the event of any inconsistency between Client's and Company's system records relating to Client's claim, unless the discrepancy is caused by any material error or malfunction of Company's system, Company's system record shall at all times prevail and be the final and conclusive reference for calculating the Service Credits to be provided to Client.

The Service Credits provided in this SLA and the applicable Order are Client's sole and exclusive remedy for any failure in the performance of Services and Company shall not be liable to Client or any person claiming through Client for any direct, indirect, consequential or incidental damages or losses or expenses whatsoever, including but not limited to, loss of profits or business and irrespective of whether the claim arises in contract, tort (including negligence), or otherwise.

[End of SLA]