MLflow Hosting in Germany
WZ-IT operates MLflow as a traceable ML platform with a tracking server, database, artifact storage, model registry and secured access.
About the technology
Manage experiments and models reproducibly
MLflow records parameters, metrics, artifacts and model versions. For teams, tracking, storage, access, retention and deployment paths must be designed together.
We assess frameworks, data volumes, CI/CD, target environments and compliance requirements and turn them into an operable architecture.
MLflow for the model lifecycle
Track experiments
Compare parameters, metrics, runtimes and results across training and evaluation runs.Manage artifacts
Store model files, charts and other results with versions in a connected artifact store.Automate approvals
Move registered models through controlled testing, approval and deployment processes.Bring teams together
Make runs and model versions centrally visible and document responsibilities transparently.Integrate tools
Connect training, CI/CD, object storage and target platforms through supported interfaces.Secure access
Protect the tracking server, artifacts, database and administrative paths from unauthorised access.Migration and takeover
Assess existing data and configuration, migrate them in a test run and move to managed operations through a controlled cutover.Identity and secure access
Secure SSO, roles, administrative paths and external access for the application and existing infrastructure.Backup and recovery
Back up all stateful components consistently and document the recovery path for the agreed scope.Updates, monitoring and operations
Monitor and update the application and its technical dependencies and operate them under the agreed service level.
Operations beyond installation
Operate MLflow for a traceable ML and AI lifecycle
Tracking, backend store, artefact storage, registry and connected training systems require consistent security and recovery.
Tracking and registry
Keep experiments, runs, models and approval stages available.Metadata and artefacts
Protect database and object storage consistently.Training and deployment
Connect CI/CD, notebooks, pipelines and serving systems.Access and operations
Control authentication, retention, monitoring and updates.
Area
Responsibility for platform and model approval
WZ-IT operates MLflow and its stores; model quality and business approval remain with customer teams.
| Area | Responsibility | Scope and boundaries |
|---|---|---|
| MLflow services | WZ-IT | Deployment, proxy, backend connection, monitoring and updates. |
| Backend and artefact store | WZ-IT | Configuration, encryption, backup and restore. |
| CI/CD and training | Shared | We integrate; the customer supplies pipelines and runtime context. |
| Model stages | Shared | Technical roles and gates implement the approved process. |
| Datasets and quality | Customer | Training, evaluation and risk remain customer-owned. |
| Automation | Optional WZ-IT service | Deployment and evaluation pipelines are scoped separately. |
Hosting & operations
How much does MLflow hosting cost?
We combine the right compute size with ongoing operations, backups, monitoring and a service level appropriate for the criticality of MLflow. High availability and recovery targets are designed separately where needed.
Hosting for MLflow in Germany: GDPR-compliant operations
- One managed standard application included
- .de domain included or your own domain or subdomain via CNAME
- German data centres with ISO 27001 and BSI C5, no US clouds
- 99.9 % standard availability per month
- Regular updates, CVE monitoring and prioritised patches for critical vulnerabilities
- Central logs with immutable retention
- Daily backup with 7-day retention
- 5 TB outbound traffic per workload and month
- 24/7 proactive monitoring
- Five service levels, 24/7 P1 response with Production or Critical
- Personal technical contact
Instance sizes and monthly prices
- XS · 2 vCPU, 2 GB RAM, 40 GB NVMe SSD
- €99.90 in total
- S · 2 vCPU, 4 GB RAM, 80 GB NVMe SSD
- €129.90 in total
- M · 4 vCPU, 8 GB RAM, 80 GB NVMe SSD
- €159.80 in total
- L · 8 vCPU, 16 GB RAM, 80 GB NVMe SSD
- €219.80 in total
- XL · 16 vCPU, 32 GB RAM, 80 GB NVMe SSD
- €329.80 in total
- Additional storage per 50 GB
- +€4.99
Straightforward provisioning within a few business days
- Bank transfer
All prices are net and exclude statutory VAT. The offers are addressed to businesses. All on-premises options
Other options
Try it free for 7 daysRequest a test instance
We set up a test instance for you, usually on the next business day. No payment details required. After seven days it is deleted unless you continue.
Request a test instance →Bring Your Own InfrastructureInstallation on Your Infrastructure
Installation on your own infrastructure · On-premise or in your cloud · Control over data location and access · Custom configuration · Complete documentation · Initial setup & configuration · Optional support and maintenance contract
Request a proposal →On-premises optionRun MLflow on hardware at your organisation
As an alternative to managed hosting in the data centre, WZ-IT provides the hardware, configures MLflow, and handles hardening, monitoring, updates, backup and technical support. Access can be limited to the internal network or enabled through VPN and existing identities.
View WZ-IT App Node →Sizing and deployment
Size MLflow by runs, artefacts and integrations
Experiment count, artefact volume, retention and automation determine platform capacity.
| Usage scenario | Technical starting point | Key factors |
|---|---|---|
| Small data-science team | Database and object-store assessment | Tracking, registry and artefacts are sized separately. |
| Many runs or large artefacts | Storage and retention design | Lifecycle and backup windows are decisive. |
| Automated training | API and load assessment | Concurrent runs and CI/CD are considered. |
| Production model approval | Governance and recovery design | Roles, audit and recovery are planned. |
- Runs, artefact volume, retention, identity and integrations are reviewed before a proposal.
Sizing and deployment
MLflow deployment models
MLflow should be close to training data, artefacts and pipelines.
Dedicated WZ-IT environment
Tracking and registry with European object storage.Your cloud
Integration with existing training and storage services.On-premises
Operate near local GPU, notebook and data systems.Hybrid
Central metadata with controlled artefact and training routes.
Target architecture
Separate metadata, artefacts and model approvals
MLflow references external artefacts and runtimes; every component needs a recovery path.
Teams and pipelines
Data scientists, CI/CD, notebooks and automated jobs.
Gateway and authentication
TLS, identity, APIs and restricted service credentials.
Roles and approvals
Access to experiments, models and production stages.
MLflow
Tracking, registry, metadata and lifecycle interfaces.
Backend store
Runs, metrics, tags and registry state.
Artefact store
Models, charts and versioned outputs.
Training and serving
External compute and deployment systems.
MLflow does not replace business model validation or organisational release controls.
Role in the overall system
Related services for MLflow
This page covers installation and the agreed platform operations for MLflow. Custom code, the access layer and special confidentiality requirements remain clearly separated responsibilities that can be added when needed.
Support custom code and integrations
Keep custom extensions, interfaces and dependencies controlled, updated and supportable.
View service Secure accessConnect users, sites and systems
Add identity-based access and private network paths as a separate operational layer.
View service Operate confidentiallyAssess Section 203 requirements separately
For professional secrecy holders, define contractual, access and infrastructure requirements in a dedicated operating model.
View serviceThree relevant adjacent paths instead of a long list of further products.
View the overall systemFrequently asked questions about MLflow hosting
Answers about tracking, artefacts, registry and operations.
- MLflow Hosting with WZ-IT starts at €99.90 net per month. It includes a dedicated size XS instance (2 vCPU, 2 GB RAM, 40 GB NVMe SSD), updates, daily backups with 7 days of retention, 24/7 monitoring, 5 TB of outbound traffic and a first response to critical incidents by the end of the next business day. Larger instances and higher service levels cost extra; the configurator shows the total price.
- The entry level is a dedicated size XS instance with 2 vCPU, 2 GB RAM, 40 GB NVMe SSD for €99.90; it suits lightweight applications as well as test and development environments. Size S with 2 vCPU, 4 GB RAM, 80 GB NVMe SSD costs €129.90. We check which size MLflow needs in production before provisioning. Size M (4 vCPU, 8 GB RAM, 80 GB NVMe SSD) costs an extra €29.90 excl. VAT, L (8 vCPU, 16 GB RAM, 80 GB NVMe SSD) an extra €89.90 excl. VAT and XL (16 vCPU, 32 GB RAM, 80 GB NVMe SSD) an extra €199.90 excl. VAT, each per month. Additional storage costs €4.99 excl. VAT per 50 GB and month.
- No. WZ-IT does not limit the number of users. How many users can work smoothly at the same time depends on usage and the chosen instance size XS to XL.
- MLflow Hosting can be cancelled monthly, with no minimum term and 30 days' notice to the end of the month.
- Germany and Switzerland are available at the same prices, in both cases without US clouds. The data centres are ISO 27001 certified; in Germany we additionally operate with a German provider whose data centres are BSI C5 attested. The Section 203 Managed Cloud for professional-secrecy holders is available in Germany only.
- We typically provide a production instance within a few business days after signing. For a trial, we set up a test instance, usually on the next business day; it runs free of charge for 7 days and without payment details.
- Yes. One .de domain is included with every hosting plan. Alternatively, connect your own domain or subdomain with a CNAME record.
- The agreed tracking and registry stack including database, artefact store, proxy, monitoring and backup.
- No. Training and GPU resources are connected or quoted separately.
- Artefact store and backend database are backed up with coordinated recovery points.
- Yes, especially when training data and GPUs are already local.
- Yes. Credentials, stages, tests and deployment targets are scoped explicitly.
Knowledge
More about MLflow
Tutorials & guides
- Self-Hosted ChatGPT: vLLM with Qwen3.5-122B on Two RTX PRO 6000 Blackwell GPUsYour own ChatGPT, running entirely in-house: a familiar chat interface, a large language model on your own hardware behind it, without a single prompt going...
- Self-Hosted Langfuse: LLM Observability and AI Logging for the EU AI ActAnyone bringing AI into production business processes quickly faces an uncomfortable question: what is actually happening in there? Which prompt went to which model, why...
- EU AI Act: High-Risk Obligations Deferred to December 2027 - What Still Starts in August 2026The EU AI Act's high-risk obligations do not start on 2 August 2026. Stand-alone high-risk systems under Annex III now have until 2 December 2027,...
Matching solutions for your project
- vLLMHigh-performance inference and serving engine for large language models with high concurrency.
- SGLangHigh-performance serving engine and runtime for large language and multimodal models.
- PostgreSQLAdvanced open-source object-relational database system
- LangfuseSelf-hosted LLM observability for tracing, prompt management, evaluations, costs and quality assurance
- DifyVisual platform for building AI applications, agents, workflows and RAG solutions.
- JupyterHubMulti-user platform for notebooks, data science, research and internal AI teams.
Reviews & projects
Client feedback and projects worldwide
WZ-IT moved our studio infrastructure from decentralised individual devices to a central platform: every site is securely connected via VPN, new devices are onboarded automatically and an entire site is provisioned from a template, without manual steps on location. What impressed me most is the breadth and depth of their knowledge: Timo and Robin are not a typical IT provider who sets up a server and leaves. The two of them think their way into highly complex infrastructure and software topics, work through every requirement we put in front of them, and build networking, provisioning and operations so that everything fits together in the end. WZ-IT is an excellent partner for complex software, network and architecture projects.
Steve KirchnerManaging Director, nextGYM GmbH
International
Built in Germany's Ruhr Valley. Running worldwide.
WZ-IT designs, develops and operates infrastructure and software for clients in Germany and internationally. We deliver projects remotely and continue supporting them in ongoing operations after go-live.
Selected projects
- LiveRemote access platform for distributed water treatment plantsABCO Water Systems · Australia · Secure Access & OT
- 81%AWS migration with substantially lower operating costsEVA Real Estate · UAE · Cloud exit & infrastructure
- 98%Provision new sites in minutes instead of hoursnextGYM GmbH · Germany · Automation & IoT
Read client reviews
- Secure your Proxmox & backup setup
- Modernize your infrastructure - sovereign
- Plan a sovereign open-source stack
- Integrate a local AI solution
- Modernize your legacy software
- Cut cloud cost - up to −81%
- Build a high-availability Proxmox cluster
- Virtualize with Managed Proxmox
- Design an open-source AI architecture
- Get collaboration fully managed
- Ship your prototype to production
- Connect sites and clusters securely
Hosting & operations
Enquire about hosting and operations for MLflow
Briefly describe the current state and objective for MLflow. We assess infrastructure, integration, and ongoing operations.
- Straight with Timo and Robin - no sales team, no pitch
- A professional assessment of your situation
- Concrete next steps for infrastructure, software or AI
The conversation is free and non-binding.
WZ-IT's advice on our Azure migration was technically sound and completely non-binding right from the intro call - we took away a great deal.
Jakob ÖschlbergerInno7 GmbH
WZ-IT's advice on our Azure migration was technically sound and completely non-binding right from the intro call - we took away a great deal.