Live on 0G Galileo · 4 pools active

A live atlas
of decentralized
compute.

Run inference on a network of independent operators. Pay by the second with on-chain streaming payments. Watch every token get computed, in real time.

Active nodes
24
Avg latency
11.2 tok/s
Settled today
8,402 USDCx
qwen-pool-1 · live networkstreaming
node-anode-bnode-cnode-dnode-enode-fnode-g◆orchestrator
node-c earning
+0.0042 USDCx/s
coalition activated
0xabc…f019 ↗

Your GPU earns nothing sitting idle. List it once, get paid by the second.

How it works

Three steps. Zero trust required.

Step 1: choose a model
01
Choose a model
Pick from open-source LLMs hosted across our operator pools. Each pool exposes its node count, price, and latency.
Step 2: set your budget
02
Set your budget
You commit USDCx upfront. A Superfluid stream opens to the operators the moment inference begins.
Step 3: watch it compute
03
Watch it compute
Tokens stream back in real time. If any node falters, KeeperHub slashes them on-chain and your refund happens automatically.

Models available now

Browse all →
Qwen3-4B
4B parameters
2 pools0.02 USDCx
Llama-3.1-8B
8B parameters
1 pool0.04 USDCx
Mistral-7B
7B parameters
1 pool0.03 USDCx
Phi-3-mini
3.8B parameters
3 pools0.018 USDCx
Common questions

Everything you'd want to know.

For node operators
For users running inference
General

Verifiable compute. Streaming payments.

Every job is settled on 0G Galileo with a public audit trail. Every operator earns by the second.