Skip to main content
ParalleliX Docs

Tutorial: CLI Command Reference

1 minTutorials

CLI command reference.

// parallelix-node v1.2.1 · nine commands · macos + linux · windows via wsl2

Every parallelix-node subcommand, one card each: synopsis, flags, what it does, the copyable command, and example output. Inference runs on Ollama; the node key signs liveness and results only, never the staking wallet.

parallelix-node probe
parallelix-node probe

Detects CPU cores, memory, and GPU (NVIDIA via nvidia-smi, Apple Silicon via Metal, AMD via ROCm), confirms Ollama is running, and prints the tier the machine maps to. CPU is tier 1; GPU is tier 2/3/4 by VRAM.

parallelix.nodenode · probe
$parallelix-node probe
cpu 16 cores
memory 64 GB
gpu NVIDIA H100 · 80 GB VRAM
ollama running (v0.x)
maps to tier 4GPU · datacenter
GPUH100VRAM80 GBTIER4// sample
probe
parallelix-node probe
parallelix-node init
parallelix-node init [--wallet 0x..] [--force]

Generates the separate, low-value secp256k1 node key at ~/.parallelix/node.key (mode 0600), writes config.json, and prints the nodeKeyHash you register on-chain. Refuses to overwrite an existing key without --force; with --force it backs the old key up before replacing it.

  • --wallet 0x.. Record the staking wallet for reference (the CLI never signs with it).
  • --force Overwrite an existing node key. Backs up the old key first, but the old node is orphaned.
parallelix.nodenode · init
$parallelix-node init
node key generated~/.parallelix/node.key · 0600
config written~/.parallelix/config.json
register on-chain (Console), using:
tier 3 (GPU · high)
nodeKeyHash0x9c27a4f1...e83b
KEYSECP256K1PERMS0600WALLETSEPARATE// sample
init
parallelix-node init
parallelix-node models
parallelix-node models [pull <name>]

Lists the models installed in the local Ollama runtime, or pulls one. The node serves inference with these models, so at least one must be present before start.

  • pull <name> Pull a model into Ollama (defaults to llama3.2). First pull downloads it. Bare `models` just lists.
parallelix.nodenode · models
$parallelix-node models pull llama3.2
ollama models (http://127.0.0.1:11434)
llama3.2open-source · ready
qwen2.5
RUNTIMEOLLAMAMODELSOSSREADY2// sample
models
parallelix-node models pull llama3.2
parallelix-node start
parallelix-node start --node-id <id> [--gpu|--cpu] [--model m]

Runs the daemon in the foreground: attaches to the on-chain node, sends signed liveness heartbeats every 10s (your uptime), polls for dispatched inference requests, runs each whole through Ollama, and returns the result with a SHA-256 Proof-of-Execution. Self-stops cleanly if the coordinator reports the node is not_staked or in_cooldown. Drains on SIGINT / SIGTERM.

  • --node-id <id> Required. The on-chain nodeId this machine serves.
  • --gpu | --cpu Force the execution mode. Defaults to gpu when a GPU is detected.
  • --model m Prefer a specific installed model. Defaults to llama3.2.
parallelix.nodenode #4217 · start
$parallelix-node start --node-id 4217 --gpu
daemon upheartbeating + serving
node onlinenodeId #4217
serving inference · llama3.2 (open-source)
proof-of-executionsha256 per result
active · heartbeats 142 · served 37
STATEACTIVEPOESHA-256REWARDSSTREAMING// sample
start
parallelix-node start --node-id 4217 --gpu
parallelix-node service
parallelix-node service --node-id <id> [--gpu|--cpu]

Installs the node as a managed background service that survives logout and reboot and auto-restarts on crash: systemd on Linux (ordered after ollama.service), launchd on macOS. On Windows, run the node inside WSL2 where the Linux path applies.

  • --node-id <id> Required. The on-chain nodeId this machine serves.
  • --gpu | --cpu Force the execution mode for the installed service.
parallelix.nodenode · systemd
$parallelix-node service --node-id 4217 --gpu
installed + startedservice 'parallelix-node'
logs: journalctl -u parallelix-node -f
stop: sudo systemctl stop parallelix-node
survives rebootauto-restart on crash
MANAGERSYSTEMDORDERAFTER OLLAMABOOTENABLED// sample
service
parallelix-node service --node-id 4217 --gpu
parallelix-node verify
parallelix-node verify

Runs diagnostics with remediation hints: Node version, node key presence and permissions, config validity, detected tier, Ollama running, models installed, and coordinator reachability. Exits non-zero if any check fails.

parallelix.nodenode · verify
$parallelix-node verify
node v22.x
node key presentperms 600
ollama runningv0.x
model llama3.2 installed
coordinator reachable
all checks passedready to start
CHECKS6 / 6EXIT0READYYES// sample
verify
parallelix-node verify
parallelix-node status
parallelix-node status

Prints the local node state from the config: node address, nodeKeyHash, tier, coordinator URL, Ollama URL, and the staking contract address.

parallelix.nodenode · status
$parallelix-node status
node address 0x4b1c…9af2
nodeKeyHash0x9c27a4f1…e83b
tier 3
coordinator https://coordinator.parallelix.io
staking CA 0x7068…82b6
TIER3COORDREACHABLECONFIGVALID// sample
status
parallelix-node status
parallelix-node logs
parallelix-node logs

Prints the last lines of the node log at ~/.parallelix/node.log: starts, served requests with PoE prefixes, heartbeat status, and exit reasons.

parallelix.nodenode · logs
$parallelix-node logs
start node=4217 mode=gpu model=llama3.2
heartbeat okseq 141
served req_a3f1ms=47 · poe=0x4f9c…
SOURCEnode.logPOEPER-REQHEARTBEATOK// sample
logs
parallelix-node logs
parallelix-node version
parallelix-node version

Prints the CLI version. The reference version documented here is 1.2.1.

parallelix.nodenode · version
$parallelix-node version
1.2.1reference build
CLI1.2.1POEv1CHANNELSTABLE// sample
version
parallelix-node version

// Where to go next

Operator end to endNode CLI · specificationTroubleshooting