Hardware Requirements
PUDA deployments usually have two kinds of machines: an AI agent host and one or more edge devices. The AI agent host runs the coding agent and connects to PUDA services. Edge devices run machine-specific adapters near the hardware.
AI Agent Host
For a Cloud API setup, use at least:
| Resource | Recommended minimum |
|---|---|
| CPU | 1 to 2 cores |
| System RAM | 4 GB |
| GPU / VRAM | Not required |
| Storage | 2 GB to 5 GB free disk space |
This is for agents that call cloud-hosted AI models. It does not include local LLM inference.
Edge Devices
Edge devices run machine-specific services such as puda-opentrons-edge. These services do not run the AI model locally; they connect to PUDA over NATS and call the machine or instrument API on the local network.
For a lightweight Python edge service, use at least:
| Resource | Recommended minimum |
|---|---|
| CPU | 1 core |
| System RAM | 1 GB |
| GPU / VRAM | Not required |
| Storage | 1 GB to 2 GB free disk space |
| Network | Reachable NATS server and reachable machine API, such as the Opentrons OT-2 IP address |
| Runtime | Linux with Python 3.14+ and uv |
Use 2 CPU cores and 2 GB RAM if the edge device will also run Docker, camera streaming, local data processing, or multiple machine adapters.