Pricing

Private inference.
Hardware that fits.

A new appliance, or the infrastructure you already own. We configure and maintain private inference around your business.

Two ways to get started.

Choose your starting point. We take care of the inference environment.

New hardware

A new appliance.

A dedicated system sized for your workloads, sourced through our distributor-backed channel and configured for your environment.

  • Workload assessment and hardware sizing
  • Hardware sourcing and system configuration
  • Deployment, validation and operational handoff
Get an appliance quote
Existing hardware

Retrofit your infrastructure.

Put your existing servers or appliances to work. We assess the hardware, prepare the environment and configure it to run our inference system.

  • Hardware and capacity suitability review
  • Installation and configuration on supported hardware
  • Validation, tuning and operational handoff
Discuss a retrofit

The same managed service.
Either way.

Your quote reflects the models and workloads you need, the infrastructure involved and the level of operational support you choose.

Configure
Model selection, inference runtime, access controls and a documented API handoff.
Maintain
Operational monitoring, agreed updates, support, and backup and recovery procedures.
Optimize
Performance tuning and capacity reviews as your workloads evolve.

What does doing it yourself cost?

The hardware is only part of it. Building and maintaining the environment takes engineering time, operational coverage and specialist expertise.

With LLM Machines, you get a defined configuration and maintenance scope, quoted for your environment.

Illustrative in-house budget €596,000 Year one

€400,000 of this is annual team cost.

Adjust the in-house estimate

Enter annual team costs including employment overheads, annual tooling costs, and year-one legal and hardware costs in euros. Already own the hardware? Set that input to zero.

Illustrative year-one scenario for a dedicated four-person team. Adjust these planning inputs for your team, region and existing infrastructure.

Cloud vs. on-prem.

Compare the infrastructure, commercial model and operational control.

Area Cloud AI APIs LLM Machines
Data controlPrompts leave your perimeterData stays inside your environment
Commercial modelUsage-based token billingCustom quote for owned infrastructure
PortabilityProvider-specific APIs and policiesDocumented API and source-available first-party Product source
DeploymentFast to start, external to your perimeterControlled deployment inside your infrastructure
AuditabilityLimited to provider exportsMetadata-only activity and audit records on-prem

Before we talk.

A few useful details for your first conversation.

What should we share for a retrofit review?

Your server or appliance specifications, GPU model and memory, system RAM, storage and network setup, plus the models and workloads you want to run. We use this to assess suitability and identify any upgrades before preparing a quote.

Can you help connect our applications?

Connect your favourite apps and harnesses through our documented inference API. For custom workflows, we can scope a separate integration engagement with a measurable outcome.

Where will the system run?

At your premises or in a data centre you contract and control. Inference runs inside your environment, on a new appliance or suitable existing hardware.

How is Product source licensed?

Original first-party LLM Machines Product source is source-available under the unmodified PolyForm Internal Use License 1.0.0. Third-party software keeps its upstream licence.

Start with what you need.

Bring a workload, a hardware specification or simply a question.

Talk to sales