Small Office Starter

Perfect for small offices—quiet, no cooling required

From £299/month

Monthly rental • Full document indexing (RAG) and LLM setup included

The Small Office Starter is perfect for small teams and offices wanting private AI without the noise or cooling concerns. Fully configured with document indexing (RAG) and LLM setup included. Runs LLaMA 3 70B for powerful AI capabilities. Quiet operation means it can sit right in your office. Ideal for document Q&A, content generation, and internal chatbots.

Request Quote Schedule Consultation

Hardware Specifications

GPU: 2 x RTX 4000 Blackwell giving 48GB VRAM

CPU: Dual Intel Xeon Gold 6138 (20 cores, 40 threads each)

RAM: 96GB DDR4 ECC

Storage: 2TB NVMe SSD

Networking: Dual Gigabit Ethernet

Power: 1450W PSU

Included Software Stack

Ollama or vLLM runtime configured
LLaMA 3 70B model preloaded
Full document indexing (RAG) setup included
ChromaDB or Qdrant vector database configured
Document ingestion pipeline ready to use
Simple web UI (React-based)
REST API for integrations

Ideal Use Cases

Internal knowledge base Q&A
Document summarisation
Email draft generation
Customer support chatbot
Code documentation assistant

Performance Expectations

Performance metrics for LLaMA 3 70B model running on this server configuration.

Metric	Single User	2-5 Users
Tokens per Second	45 tokens/sec	35 tokens/sec
Response Time	2.2 seconds	3.5 seconds
Concurrent Requests	1 request	5 requests

Delivery & Support

Available for immediate rental. Full document indexing (RAG) and LLM setup included. Includes remote setup assistance and 1-hour onboarding call. Quiet operation—no special cooling required.