Agent-native cloud.

European infrastructure for AI builders - compute, databases, and full AI stacks, billed by the hour.

Start free - £200 creditsView pricing
01

Infrastructure your agents run

Point your agent at a project. It provisions real EU infrastructure and hands you a live URL.

02

Full cloud stack, by the hour

Real VMs, managed databases, autoscaling clusters, and private networks. EU-hosted.

03

Self-healing deploys

Builds retry, wait out locks, and recover from the failures that break most deploys.

redu MCP

Infrastructure your agents run.

Point your agent at a project through the redu MCP. It provisions what the project needs, real VMs, a managed database, and autoscaling clusters under load, runs it in the EU, and gives you a live URL.

Your agent works inside your own account and quota, on real infrastructure, not a sandbox, and you see every step.

How the MCP works →Watch real deploys →
# 1. Add redu to your agent
claude mcp add --transport http \
  redu-cloud https://mcp.redu.cloud/mcp

# 2. From your project folder, run:
"Deploy this project to redu"

# You get a plan ready. You approve it.
# It deploys, and self-heals complicated workloads.
Infrastructure

Full cloud stack. By the hour.

Our own infrastructure, not a reseller. Real machines in a European datacenter, with EU data residency. Use the console, the API, or your agent. No commitment, no minimums. £200 free credits to start. Full pricing →

Compute instancesAutoscaling clustersPrivate networksManaged PostgreSQLBlock storage
Autoscaling clusterskeyHorizontal scaling, policy-based, spin up or down automatically under load
Compute instances8 vCPU · 16 GB RAM · from £0.0972/hr, billed to the minute
Load balancersTCP + HTTP, distribute traffic across your fleet automatically
Private networksIsolated VLAN per project, custom CIDR, no cross-tenant routing
Managed PostgreSQLDedicated VM, automated backups, connection pooling ready
Managed RedisPersistent, in-memory, hourly billed, no shared instances
Block storageAttach to any VM, from £0.0001/GB/hr
Backups & snapshotsPoint-in-time, manual or scheduled, stored separately
Autocoding agents

A fleet that ships code while you sleep.

A coordinator assigns tasks to a fleet of agents. Each writes code and opens a PR. You wake up to merged features.

Agents have full access to your infrastructure. They can spin up VMs, volumes, and clusters if the task needs it. Not a sandbox.

Full setup guide →
Billing globalBillingRecorder activity tests#209 · merged Jun 3
Fix Redis and Qdrant instance_id collision risk#208 · merged Jun 3
Add created_at to qdrant_instances list response#204 · merged Jun 2
Keypair delete confirmation modal#201 · merged Jun 2
Pagination for snapshots list#200 · merged Jun 2
Instance search by name#198 · merged Jun 2
Managed Vector DB (Qdrant)#197 · merged Jun 2
Zod validation for audit log query params#196 · merged Jun 2
Credit expiry warning flag#195 · merged Jun 2
Expose rate limit remaining header#194 · merged Jun 2
Add description field to API keys#193 · merged Jun 2
Tighten volume size limit to 2048 GB#192 · merged Jun 2
Billing dashboard resource type breakdown#191 · merged Jun 2
Add last_used_at to API keys#190 · merged Jun 2
Tests for pushHourlyUsageToStripe#189 · merged Jun 2
Pricing rules admin UI#188 · merged Jun 2
Per-user API rate limiting#186 · merged Jun 2
reduOS

The AI stack your agents run on.

One VM. Every layer pre-wired. Deploy in five minutes and give your agents a full environment to work in.

Learn more about reduOS →View on GitHub →
ModelsLiteLLM · Ollama · LangGraphModel routing, local inference, agent workflows
MemoryQdrant · SupabaseVector search and persistent storage
Automationn8n · ActivepiecesWorkflow triggers and event handling
ObservabilityLangfuse · GlitchTip · Uptime Kuma · UmamiTraces, errors, uptime, and analytics

Point your agent at redu.

£200 in free credits. No commitment.

Create your account