99% revenue sharegoogle/gemma-4-26b-a4b-nvfp4$0.03 / $0.20 per 1M
Install a node with one command
The installer checks your GPU (nvidia-smi, ≥ 32 GB VRAM), sets up Docker + the NVIDIA container toolkit, joins the SeedInfer mesh network and starts the inference agent. Your node sends a heartbeat every 30 s and is auto-verified after 2 heartbeats.
1 · Recommended
curl -fsSL https://seedinfer.com/install.sh | bash -s -- --authkey YOUR_AUTHKEY
Replace YOUR_AUTHKEY with a key generated on Become a provider.
2 · Auto-fetch key (needs jq)
curl -fsSL https://seedinfer.com/install.sh | bash -s -- --authkey $(curl -s https://seedinfer.com/api/v1/auth/request | jq -r .authkey)
Requests an auth key and installs in one go — handy for scripted setups.
3 · All options
curl -fsSL https://seedinfer.com/install.sh | bash -s -- --authkey YOUR_AUTHKEY --model google/gemma-4-26b-a4b-nvfp4 --gateway https://seedinfer.com --hostname my-gpu-node
Requirements
- Model served
- Gemma 4 26B A4B NVFP4 · 256K ctx
- GPU
- NVIDIA, ≥ 32 GB VRAM (reference: RTX 5090 32GB)
- OS
- Ubuntu 24.04 LTS
- Driver / CUDA
- 580.65+ / 13.x
- Runtime
- Docker 24+ with nvidia-container-toolkit
- Disk
- ≥ 60 GB free (model cache)
- Ports (host)
- 47900 (vLLM) · 47901 (agent)
- Network
- Outbound UDP 41641 (WireGuard mesh)
- Payout wallet
- USDC on Base (EVM 0x… address)
Verify after install
curl -fsS http://127.0.0.1:47901/health | jq
curl -fsS http://127.0.0.1:47900/v1/models | jq
curl http://127.0.0.1:47901/v1/chat/completions \
-H "Content-Type: application/json" \
-d '{"model":"google/gemma-4-26b-a4b-nvfp4","messages":[{"role":"user","content":"ping"}],"max_tokens":32}'Heartbeat every 30 s · auto-verified after 2 heartbeats.
Registry
Network nodes
Verified nodes serve traffic; pending and verifying nodes are shown dimmed.
0 verified · 0 totalJSON