System Requirements#

Hardware Requirements#

Component	Minimum	Recommended
CPU	4 cores	8+ cores
RAM	8 GB	16 GB
Disk	10 GB free	20 GB free
GPU	(optional) Intel® Arc™ / Flex / Data Center GPU	Intel® Arc™ A770 or equivalent

Note

The default LLM (Phi-4-mini-instruct-int4-ov) can run on either GPU or CPU. Set TARGET_DEVICE=CPU in your environment to use CPU-only inference. GPU provides significantly lower latency and is recommended for production workloads with high alert frequency.

Software Requirements#

Software	Version	Notes
Docker Engine	24.0 or later	Install Docker
Docker Compose	v2 plugin	Bundled with Docker Desktop; install separately on Linux
Python	3.12 or later	Required only for running tests locally
`uv`	Latest	Required only for local development; `pip install uv`

Network Requirements#

The following network ports are used by default:

Port	Service	Configurable via
`8000`	Alert Agent Service REST API	`PORT` environment variable
`9001`	OVMS LLM server (host-mapped)	`LLM_PORT` environment variable
`1883`	MQTT broker (host-mapped)	`MQTT_HOST_PORT` environment variable

Within the Compose network, OVMS listens on port 9000 and the bundled MQTT broker listens on port 1883. LLM_PORT and MQTT_HOST_PORT only control how those ports are published on the host. MQTT_PORT specifies the destination port used by the Alert Agent Service when connecting to an MQTT broker.

Proxy Settings#

If your environment requires an HTTP proxy, set the standard proxy environment variables before starting the service:

export http_proxy=http://proxy.example.com:8080
export https_proxy=http://proxy.example.com:8080
export no_proxy=localhost,127.0.0.1,ovms-llm

These are forwarded automatically to both containers in the Docker Compose setup.