For advanced
Self-hosting: step-by-step guide
At Gonka, GPUs process network neural network requests and receive GNK tokens for each completed request.
Don't want to deal with servers?
You don't need your own equipment to earn GNK. Join a pool — invest from $30 and get GNK without technical skills.
Hardware Requirements
Network Node (network server)
- 16-core CPU (amd64)
- 64GB+ RAM
- 1TB NVMe SSD
- 100Mbps+ network
ML Node (compute node)
- MLNode: MiniMax M2.7 — 3.0.14 or newer, DeepSeek V4 Flash — 3.0.16 or newer, GLM-5.3-Flash — 3.1.0 (not 3.0.17)
- Go 1.22.8, Docker Desktop 4.37+, Java 19+
- NVIDIA GPU (newer generation than Tesla)
- Minimum 320GB VRAM total per MLNode container
- You can mix and match GPUs in a node — as long as they can fit the models approved by governance and pass PoC
- Reference configuration: 8xH200 per node
- NVIDIA Container Toolkit, CUDA Toolkit 12.6-12.9 (check your version with nvidia-smi); for GLM-5.3-Flash — CUDA driver 13 (580+)
Tip: No GPUs of your own? Rent bare-metal from GPU providers from ~$11,600/month for 8xH100.
Gonka.Top — the best referral program in the ecosystem
Collective plan from $100. 25% L1 + 5% L2 of pool profits. Monthly payouts for 180 days.
Setup Steps
01 Install inferenced CLI
Download the CLI archive for your OS (v0.2.16 or newer) from GitHub releases.
# Download and install inferenced CLI
wget https://github.com/gonka-ai/gonka/releases/download/release/v0.2.16-post1/inferenced-linux-amd64.zip
unzip inferenced-linux-amd64.zip
chmod +x inferenced
sudo mv inferenced /usr/local/bin/inferenced02 Create a cold wallet key
The cold wallet is used to manage the node and receive rewards.
inferenced keys add cold-walletSave your mnemonic phrase in a safe place!
03 Clone the repository and configure
git clone https://github.com/gonka-ai/gonka.git -b main
cd gonka/deploy/join
cp config.env.template config.env
# Edit config.env — enter your parameters04 Download model weights
Active network models — MoE FP8 (currently MiniMax M2.7, DeepSeek V4 Flash, and GLM-5.3 Flash; the set is defined by governance). Weight loading via HuggingFace CLI. Learn more →
pip install -U "huggingface_hub[cli]"
# use the current model id from the network config / governance (example: MiniMax M2.7)
huggingface-cli download MiniMaxAI/MiniMax-M2.705 Start Docker services
docker compose -f docker-compose.yml -f docker-compose.mlnode.yml pull
source config.env && docker compose up tmkms node -d --no-deps06 Register on-chain
On the server, create an operational (warm) key and register the host, then on the local machine where the cold key resides, grant the warm key rights for service transactions.
# On the server, in the api container: create an operational (warm) key
docker compose run --rm --no-deps -it api /bin/sh
printf '%s\n%s\n' "$KEYRING_PASSWORD" "$KEYRING_PASSWORD" | inferenced keys add "$KEY_NAME" --keyring-backend file
# Same place: register the host on the network
inferenced register-new-participant \
$DAPI_API__PUBLIC_URL \
$ACCOUNT_PUBKEY \
--node-address $DAPI_CHAIN_NODE__SEED_API_URL
# On the local machine: grant rights to the warm key (fee 0.002 GNK)
inferenced tx inference grant-ml-ops-permissions \
cold-wallet <warm-key-address> \
--from cold-wallet \
--gas 2000000 --gas-prices 1ngonka \
--node <seed-api-url>/chain-rpc/Since October 2026 (network update v0.2.16), granting rights is paid: without --gas-prices the network will reject the transaction with an insufficient fee error. With --gas 2000000 and a price of 1 ngonka per gas, the fee is 0.002 GNK. Keep a reserve of GNK in your cold wallet: the warm key's fees are also deducted from it — the official guide recommends starting with 10 GNK.
07 Configure SSL and launch
Configure DNS (Cloudflare/AWS/GCP/Azure/DO/Hetzner), obtain an SSL certificate, launch the full network, and verify registration.
source config.env && \
docker compose -f docker-compose.yml -f docker-compose.mlnode.yml up -d
# Check node status
curl http://<PUBLIC_IP>:8000/chain-rpc/status
# Verify registration
curl http://node2.gonka.ai:8000/v2/participants/<cold-wallet-address>Don't want to set it up yourself?
There are GPU hosting and pools that do everything for you:
Dedicated server
- Gonka.Top Dedicated — Gonka.Top — ~$12,000/month, full turnkey service, 100% of the power is yours
Mining pools
- Gonka.Top — from $100, 25% referral + 5% L2, monthly payouts
- GonkaPool.ai — Telegram bot, convenient entry
- CloudMine (Mingles) — contracts by "weight" 50–10000, payment USDC on Base
- Gonka.Wallet — Telegram Mini App, one-click wallet, built-in pools
Best referral: Gonka.Top — 25% of profits. Lowest entry point: Ancapex — from $1, daily payouts.
What Affects Profitability
- Weight: base weight = 20% automatically. The remaining 80% is activated by staking GNK as collateral. Learn more →
- GPU performance: the more powerful the graphics card, the more neural network requests are processed per round.
- Network utilization: at 40-60% load the price is stable. Above 60% — price increases (more revenue per request).
- Fair rewards: The system automatically verifies each computation (<10ms) and fairly distributes rewards. During network updates, there is a transition period (~5 hours) so that no one loses income. Learn more →