Joining the HART OS Compute Network¶

How to join the HART OS compute network as a regional compute provider.

Prerequisites¶

A running HART OS instance (python hart_intelligence_entry.py on port 6777)
At least one GPU (recommended) or CPU-only node
Network connectivity to at least one seed peer

Step 1: Install and Configure¶

# Clone and install
git clone <repo-url> && cd HARTOS
python3.10 -m venv venv310
source venv310/Scripts/activate
pip install -r requirements.txt

Set environment variables:

export HEVOLVE_NODE_ID="my-node-001"
export HEVOLVE_NODE_TIER="regional"         # flat | regional | central
export HEVOLVE_BASE_URL="https://my-node.example.com:6777"

Step 2: Configure Cause Alignment¶

Every provider declares why they are contributing compute. This is gossipped to the network so peers know your motivation.

Supported causes:

Cause	Description
`democratize_compute`	Making AI compute accessible to everyone (default)
`frontier_training`	Training the collective HiveMind model
`thought_experiments`	Running community-proposed thought experiments

Step 3: Join the Network¶

All examples below use https://hevolve.ai. Replace with http://localhost:6777 if self-hosted.

curl -X POST https://hevolve.ai/api/settings/compute/provider/join \
  -H "Content-Type: application/json" \
  -d '{
    "cause_alignment": "democratize_compute",
    "electricity_rate_kwh": 0.12,
    "offered_gpu_hours_per_day": 8,
    "compute_policy": "local_preferred"
  }'

Response:

{
  "joined": true,
  "node_id": "my-node-001",
  "config": {
    "compute_policy": "local_preferred",
    "max_hive_gpu_pct": 50,
    "allow_metered_for_hive": false,
    "offered_gpu_hours_per_day": 8,
    "accept_thought_experiments": true,
    "accept_frontier_training": false
  }
}

This creates a NodeComputeConfig row and updates your PeerNode identity.

Step 4: Verify Provider Status¶

curl https://hevolve.ai/api/settings/compute/provider

Returns your contribution score, compute stats, pending settlements, and reward summary. See dashboard.md for full details.

Step 5: Peer Discovery¶

Once joined, the gossip protocol automatically discovers and connects to other nodes. Seed peers are configured via HEVOLVE_SEED_PEERS environment variable. Your cause alignment is included in gossip messages so the network knows your contribution intent.

What Happens Next¶

Your node begins accepting hive tasks based on your compute_policy
GPU hours, inferences, and energy are tracked on your PeerNode record
Metered API costs are recorded as MeteredAPIUsage entries
Settlements are processed automatically (see settlement.md)
Your contribution score updates every ~50 seconds via aggregate_compute_stats()