Fluxus
AI Private Custom Model Hosting
AI Private Custom Model Hosting. Elastic serverless AWS-based hosting for your private models accessible by secure API with key management, rate limiting, caching, geo-limiting.
Features
- Fully private
- Price by usage
- Self-managed or managed by us
- Rate-limiting available
- Can be geo-constrained
Benefits
- Custom models based on your data
- High performance
- Cost effective
- Secure and private, no shared data in or out
- Low/no maintenance
Pricing
£0.50 to £0.50 a transaction
- Free trial available
Service documents
Request an accessible format
If you use assistive technology (such as a screen reader) and need versions of these documents in a more accessible format,
email the supplier at info@fluxus.io.
Tell them what format you need. It will help if you say what assistive technology you use.
Framework
G-Cloud 14
Service ID
9 4 4 4 8 9 4 9 6 3 2 8 8 1 9
Contact
Fluxus
Django Beatty
Telephone: 01392581040
Email: info@fluxus.io
Service scope
- Service constraints
- AWS-based
- System requirements
- None
User support
- Email or online ticketing support
- Email or online ticketing
- Support response times
- Within working day
- User can manage status and priority of support tickets
- No
- Phone support
- No
- Web chat support
- No
- Onsite support
- Yes, at extra cost
- Support levels
-
Standard support is by email during UK office hours Mon-Fri.
UK-based TAM available on request, on site or remote.
If support is critical we have a 3rd party partner based in EU who can provide fully technical support at various levels up to 24/7 by phone. - Support available to third parties
- Yes
Onboarding and offboarding
- Getting started
- Online training included.
- Service documentation
- Yes
- Documentation formats
- Other
- Other documentation formats
- OpenAPI
- End-of-contract data extraction
- On request or self serve.
- End-of-contract process
- Account is wiped as standard, but can be archived on request. Models and any associated data can be exported at any point during contract, either self-serve or as managed service, depending on requirements.
Using the service
- Web browser interface
- No
- API
- Yes
- What users can and can't do using the API
- Managed or self-service depending on requirements. Services delivered via AWS API Gateway. Auth via Cognito unless other requested.
- API automation tools
-
- Ansible
- Chef
- OpenStack
- SaltStack
- Terraform
- Puppet
- Other
- API documentation
- Yes
- API documentation formats
- Open API (also known as Swagger)
- Command line interface
- No
Scaling
- Scaling available
- Yes
- Scaling type
- Automatic
- Independence of resources
- Serverless architecture means each user has independent resources.
- Usage notifications
- Yes
- Usage reporting
-
- API
Analytics
- Infrastructure or application metrics
- Yes
- Metrics types
-
- HTTP request and response status
- Other
- Other metrics
-
- Activity
- Location
- Number of users
- Token usage
- Query response speed
- Performance against quality benchmarks
- Reporting types
- Real-time dashboards
Resellers
- Supplier type
- Not a reseller
Staff security
- Staff security clearance
- Other security clearance
- Government security clearance
- Up to Baseline Personnel Security Standard (BPSS)
Asset protection
- Knowledge of data storage and processing locations
- Yes
- Data storage and processing locations
-
- United Kingdom
- European Economic Area (EEA)
- Other locations
- User control over data storage and processing locations
- Yes
- Datacentre security standards
- Complies with a recognised standard (for example CSA CCM version 3.0)
- Penetration testing frequency
- Less than once a year
- Penetration testing approach
- In-house
- Protecting data at rest
-
- Physical access control, complying with CSA CCM v3.0
- Physical access control, complying with SSAE-16 / ISAE 3402
- Encryption of all physical media
- Data sanitisation process
- Yes
- Data sanitisation type
-
- Explicit overwriting of storage before reallocation
- Deleted data can’t be directly accessed
- Equipment disposal approach
- A third-party destruction service
Backup and recovery
- Backup and recovery
- Yes
- What’s backed up
-
- Models
- Stored prompts
- System user data
- Fine tune data
- Embedded (RAG) data
- Backup controls
- Users can specify whether to back up prompts, fine-tune, embedded (RAG) data.
- Datacentre setup
- Multiple datacentres with disaster recovery
- Scheduling backups
- Supplier controls the whole backup schedule
- Backup recovery
-
- Users can recover backups themselves, for example through a web interface
- Users contact the support team
Data-in-transit protection
- Data protection between buyer and supplier networks
-
- Private network or public sector network
- TLS (version 1.2 or above)
- IPsec or TLS VPN gateway
- Data protection within supplier network
-
- TLS (version 1.2 or above)
- IPsec or TLS VPN gateway
Availability and resilience
- Guaranteed availability
-
AWS standard for API Gateway and Lambda. Currently:
Monthly Uptime Percentage
Less than 99.95% but greater than or equal to 99.0%
Service Credit Percentage
10%
Monthly Uptime Percentage
Less than 99.0% but greater than or equal to 95.0%
Service Credit Percentage
25%
Monthly Uptime Percentage
Less than 95.0%
Service Credit Percentage
100% - Approach to resilience
- Available on request. (12-factor architecture adherent to AWS well-architected framework.)
- Outage reporting
- Dashboard and email alerts
Identity and authentication
- User authentication
-
- 2-factor authentication
- Identity federation with existing provider (for example Google apps)
- Access restrictions in management interfaces and support channels
- Access to management interfaces and support channels is restricted to authorized personnel only. We follow the principle of least privilege, granting access based on job roles and responsibilities. Multi-factor authentication is enforced for all user accounts. Access rights are regularly reviewed and updated to ensure they remain appropriate. Support channels, such as email and chat systems, are secured and monitored. Communication through these channels is encrypted to protect sensitive information. We maintain strict access control policies and procedures, which are regularly audited to ensure compliance and effectiveness.
- Access restriction testing frequency
- At least every 6 months
- Management access authentication
-
- 2-factor authentication
- Identity federation with existing provider (for example Google Apps)
- Devices users manage the service through
- Directly from any device which may also be used for normal business (for example web browsing or viewing external email)
Audit information for users
- Access to user activity audit information
- Users contact the support team to get audit information
- How long user audit data is stored for
- User-defined
- Access to supplier activity audit information
- Users contact the support team to get audit information
- How long supplier audit data is stored for
- At least 12 months
- How long system logs are stored for
- At least 12 months
Standards and certifications
- ISO/IEC 27001 certification
- No
- ISO 28000:2007 certification
- No
- CSA STAR certification
- No
- PCI certification
- No
- Cyber essentials
- No
- Cyber essentials plus
- No
- Other security certifications
- No
Security governance
- Named board-level person responsible for service security
- Yes
- Security governance certified
- No
- Security governance approach
-
Principles:
Customer Data Ownership
Data Quality
Data Security
Data Sharing (GDPR compliant)
Consistent Data Definitions
Roles and Responsibilities:
Data Governance Committee: Policy enforcement and issue resolution
Employees: Preserve customer data security, integrity, and privacy - Information security policies and processes
-
We prioritize data security and privacy with these key processes:
Secure development: Implement secure coding, encryption, and regular updates.
Access control: Use least privilege principle, strong authentication, and regular permission reviews.
Data backup and disaster recovery: Regularly backup data and establish a recovery plan.
Incident response: Develop a plan and designate a team to manage security incidents.
Employee training: Train staff on data security and GDPR requirements.
Vendor management: Vet and monitor third-party vendors for GDPR compliance and security.
Data Protection Impact Assessments: Conduct DPIAs for high-risk processing activities.
Consent management: Obtain explicit customer consent and provide preference management.
Data retention and deletion: Establish retention policies and secure deletion processes.
Regular audits and assessments: Perform security audits and risk assessments to identify and address vulnerabilities.
Operational security
- Configuration and change management standard
- Supplier-defined controls
- Configuration and change management approach
-
Maintain a centralized inventory of all service components, including AWS resources, models, and configurations.
Use version control and unique identifiers to track changes throughout the component lifecycle.
Implement a formal change management process with documented procedures for proposing, reviewing, approving, and implementing changes.
Conduct security impact assessments for all proposed changes, evaluating risks and potential vulnerabilities.
Use infrastructure as code (IaC) and automated deployment pipelines to ensure consistent and auditable changes.
Regularly review and update the inventory and change management processes to maintain accuracy and effectiveness. - Vulnerability management type
- Supplier-defined controls
- Vulnerability management approach
- We prioritize the security of data in our hosted custom private LLMs. We leverage the fully managed AWS serverless platform and adhere to strict vulnerability management processes. These include secure configuration management, continuous monitoring using AWS native security services, and regular updates of dependencies. We implement rigorous access control measures and encrypt all data at rest and in transit. We assess the security practices of integrated third-party services and have a comprehensive incident response plan. By focusing on these key areas, we ensure the ongoing protection of data within our serverless environment.
- Protective monitoring type
- Supplier-defined controls
- Protective monitoring approach
- We employ protective monitoring processes to safeguard data in its hosted custom private LLMs. AWS security services continuously monitor the serverless environment, automatically identifying potential compromises in real-time. Upon detection, the incident response team is alerted and initiates an investigation following a defined plan. The team contains the issue, assesses the impact, and implements mitigation measures. Critical incidents are responded to within 60 minutes, with the team working to resolve the issue and restore data security. The company prioritizes swift action and effective communication throughout the process to minimize potential harm and keep stakeholders informed.
- Incident management type
- Supplier-defined controls
- Incident management approach
- Pre-defined procedures are in place for common events, ensuring a consistent and efficient response. Users can report incidents via email, which is monitored by the incident response team. The team assesses the incident, determines its severity, and follows the appropriate response plan. Throughout the incident lifecycle, the team provides regular updates to stakeholders via email. Once the incident is resolved, a comprehensive report is generated, detailing the cause, impact, and remediation steps taken. The report is shared with relevant parties for transparency and future prevention.
Secure development
- Approach to secure software development best practice
- Conforms to a recognised standard, but self-assessed
Separation between users
- Virtualisation technology used to keep applications and users sharing the same infrastructure apart
- No
Energy efficiency
- Energy-efficient datacentres
- No
Social Value
- Social Value
-
Social Value
Equal opportunityEqual opportunity
Our service promotes equal opportunity by supporting the development of inclusive and accessible language-based applications. By enabling organizations to create applications that cater to diverse populations, including those with disabilities or limited English proficiency, we help reduce the disability employment gap and tackle workforce inequality. Our platform empowers organizations to develop applications that provide equal access to information, services, and opportunities, promoting digital inclusion and reducing barriers to participation in the digital economy.
Pricing
- Price
- £0.50 to £0.50 a transaction
- Discount for educational organisations
- No
- Free trial available
- Yes
- Description of free trial
- Free service for the purpose of building PoC's and during development.
Service documents
Request an accessible format
If you use assistive technology (such as a screen reader) and need versions of these documents in a more accessible format,
email the supplier at info@fluxus.io.
Tell them what format you need. It will help if you say what assistive technology you use.