Self-hosted AI contract review for firms that can’t send data to the cloud.
VP Arbiter Enterprise runs entirely on your infrastructure. Your contracts, your AI keys, your data — never leaves your network. Built for UK firms with strict data-sovereignty, AML, or compliance requirements.
From £4,995 · One-off licence
Optional annual support contract from £1,200/year. Custom enterprise terms, SLAs, and source escrow available. Quote provided after a 20-minute call.
What Enterprise gives you
On Your Infrastructure
Deploy on your servers, your VPC, or your private cloud. Contract text never leaves your network.
Bring Your Own AI
Use your existing Anthropic, OpenAI, or self-hosted Ollama deployment. Choose models that match your compliance posture.
No Per-Seat Licensing
One licence covers your firm. Scale users without recurring per-seat costs.
UK Data Residency
Hosted in UK or EU on your terms. GDPR-aligned by architecture, not by promise.
AML & Compliance Ready
Audit logging, retention controls, and exportable analysis records to support your AML and ISO 27001 obligations.
Source Code Escrow
Optional code escrow available for firms requiring continuity guarantees.
Who Enterprise is for
- 01
UK law firms handling client contracts subject to confidentiality undertakings.
- 02
In-house legal teams at FCA-regulated entities with documented data-sovereignty obligations.
- 03
Property firms reviewing commercial leases at scale where speed and audit-trail matter.
- 04
Procurement teams running supply-chain reviews on sensitive vendor agreements.
- 05
Anyone with an internal AML, ISO 27001, or SOC 2 obligation that prevents sending text to public AI APIs.
Frequently asked questions
How does pricing work?
One-off licence from £4,995 with optional annual support and updates. Pricing scales with the number of analysts using the system, not with the number of analyses run. We provide a fixed quote after a 20-minute scoping call.
What infrastructure do we need?
A modest Linux server or container host with 16GB RAM and a GPU is sufficient for most firms. We can deploy to AWS, Azure, GCP, on-premises, or your existing Kubernetes cluster. Ollama-based deployments run entirely without an internet connection.
How long does deployment take?
Typical deployment is one working day, including infrastructure provisioning, AI model setup, and configuration. We provide written runbooks and 30 days of post-deployment support standard.
Can we modify the system?
Source code is provided. You can adapt the prompts, the report templates, and the UI to match your firm's house style. We retain ownership of the underlying engine.
What ongoing support is included?
30 days of post-deployment support is standard. Annual support contracts cover updates, AI model upgrades, and priority response. Custom SLAs available.
Next step
Book a 20-minute scoping call
We’ll discuss your infrastructure, AI model preferences, and compliance requirements. You’ll receive a fixed-price quote within one business day.
Contact us →