Skip to main content
RecordEngine runs entirely on your own hardware. This page covers everything you need before installation.

​
Hardware

​
GPU (Required)

RecordEngine requires an NVIDIA GPU with at least 16 GB VRAM. The GPU runs all AI inference locally — document extraction, summarisation, confidence scoring, and chat all happen on the GPU.
GPUVRAMStatus
NVIDIA RTX 509032 GB✅ Recommended
NVIDIA RTX 409024 GB✅ Supported — standard production config
NVIDIA RTX 3090 / 3090 Ti24 GB✅ Supported
NVIDIA A10G (AWS g5.xlarge)24 GB✅ Supported — cloud deployment
NVIDIA RTX 408016 GB⚠️ Minimum — limited headroom
AMD GPUs—❌ Not supported
Apple Silicon—❌ Not supported
RecordEngine will not run without a supported NVIDIA GPU. CPU-only operation is not supported — inference speed on CPU is 20–50x slower than GPU and not viable for production use.

​
CPU, RAM, and Storage

ComponentMinimumRecommended
CPU8 cores16+ cores
RAM32 GB64 GB
Storage200 GB SSD500 GB+ NVMe SSD
Network100 Mbps LAN1 Gbps LAN
Storage grows with your document archive. Plan for approximately 1–5 MB per document including the original file and extracted data. At 500 documents per month, 200 GB lasts several years — but NVMe SSD is strongly recommended over SATA SSD for faster model loading on startup.

​
Software

​
Operating System

OSStatus
Ubuntu 24.04 LTS✅ Recommended
Ubuntu 22.04 LTS✅ Supported
Debian 12✅ Supported
Windows Server 2022⚠️ Community support only
macOS❌ Not supported for production

​
Required Software

SoftwareMinimum versionNotes
Docker Engine24.0+
Docker Compose2.20+Included with Docker Desktop
NVIDIA GPU Driver525+Must be installed before Docker
NVIDIA Container ToolkitLatestRequired for GPU access inside containers
All other dependencies (Python, FastAPI, Streamlit, Ollama, etc.) are included inside the Docker containers — nothing needs to be installed separately on the host OS.

​
Network

​
Inbound Ports

PortServiceRequired
80HTTP (redirects to HTTPS)Yes
443HTTPS (Nginx reverse proxy)Yes
22SSH (admin access)Yes
Ports 8501 (Streamlit) and 8510 (FastAPI) should not be exposed publicly — all external access goes through Nginx on ports 80/443.

​
Internet Access

RecordEngine operates fully offline after initial setup. Internet access is only required for:
  • Initial setup: Downloading Docker images and the AI model (~7–10 GB total)
  • Software updates: Pulling new code and model versions (infrequent, on your schedule)
Once deployed, the system operates with no outbound internet connections. Documents never leave your server.

​
Firewall

If your server is behind a corporate firewall, the following outbound connections are needed only during setup:
DestinationPortPurpose
hub.docker.com443Docker image downloads
registry-1.docker.io443Docker image downloads
huggingface.co443AI model download (via Ollama)
After initial setup, all of these can be blocked if your security policy requires it.

Standard Office Deployment

Lenovo ThinkStation or Dell Precision
  • CPU: Intel Core i9 or Xeon W
  • GPU: NVIDIA RTX 4090 (24 GB)
  • RAM: 64 GB DDR5
  • Storage: 1 TB NVMe SSD
  • OS: Ubuntu 24.04 LTS
Handles 100–300 documents per day comfortably.

High-Volume Deployment

AWS g5.xlarge or equivalent
  • GPU: NVIDIA A10G (24 GB)
  • vCPU: 4
  • RAM: 16 GB
  • Storage: 500 GB+ EBS gp3
Cloud deployment for distributed teams. See Docker on Linux.

​
Client Requirements

Users access RecordEngine through a web browser — no software installation is required on client machines.
BrowserStatus
Chrome 100+✅ Recommended
Firefox 100+✅ Supported
Safari 16+✅ Supported
Edge 100+✅ Supported
The mobile chat interface works on any modern mobile browser (iOS Safari, Android Chrome).