WorkerConsoleExplorerStakeLeaderboardGovernanceDocsDownload
DEVNET

Worker

Run inference on the network and earn $LOCK. Choose browser, desktop, or native worker for your setup.

Browser Worker
WebLLM · Llama 3.2 · in-tab via WebGPU
Not Ready

Runs entirely in your browser — no install. Uses WebGPU to serve jobs while this tab stays open.

Connect wallet to start.
HARDWARE
GPU
Detecting…
VRAM
—
RECOMMENDED
Desktop Worker
Gridlock Worker app · v0.1.3

Install the desktop app for earnings, live jobs, and GPU settings. Includes the worker daemon — no terminal required.

DOWNLOAD
  1. Install the app for your operating system
  2. Open Gridlock Worker and connect your wallet
  3. Click Start — jobs arrive automatically
REQUIRES
NVIDIA GPU · CUDA
JOBS VIA
WebSocket

Server or headless setup? Use Native Worker instead.

RECOMMENDED
Native Worker
CLI · Ollama / vLLM · v0.1.3

Headless worker for servers and power users.
Runs in your terminal, connects to the router, and forwards jobs to Ollama or vLLM on your GPU.

  1. 1. Install Ollama (recommended on Windows) — the worker can auto-start it
  2. 2. In the worker folder: run setup once, then start the worker
  3. 3. Connects to the router automatically
SETUP (ONCE)
npm install && npm run build && npm link
INFERENCE (OPTIONAL — AUTO-PULLED)
ollama pull llama3.1:8b
START WORKER
gridlock-native-worker --wallet YOUR_WALLET_ADDRESS
CONFIDENTIAL / TEE JOBS (DEV OR H100)
$env:GRIDLOCK_TEE_CAPABLE="true"; gridlock-native-worker --wallet YOUR_WALLET_ADDRESS

Serves privacy jobs from the Console. Consumer RTX uses dev attestation; production needs H100 Confidential Computing.

REQUIRES
Node 18+ · NVIDIA
PACKAGE
@gridlock/native-worker

Want a UI? Use Desktop Worker instead — same network, worker dashboard included.