Network Online (0 Nodes)|⚡ P2P INFERENCE GRID ACTIVE99.98% UPTIME🌐 LATENCY: 9.8 ms SUB-SECOND RPC💎 PROOF-OF-INFERENCE PAYOUT RATE: 100% AUTOMATED TC
System Explainer

How the
Whole System
Works

You share your GPU, you earn Tensor Credits. Those credits let you use AI — even when your machine is off because someone else is doing the compute.

Live Credit Cycle

Your GPU is ON
OriginLink running
+TC
Others' requests → your GPU → you earn credits
Your TC Wallet
💎 120 TC
≈ 100M Input + 100M Output Tokens (8B LLM)
Your API calls → someone else's GPU → TC deducted
Your GPU is OFF
AI still works via P2P
−TC
Contribute more than you consume → effectively free ✓
THE SIMPLE IDEA

Your GPU earns credits. Those credits buy you AI.

Think of it like a power grid. When your laptop is on, you contribute electricity and the meter runs in your favour. When your laptop is off, you draw from the grid—paid for by what you already earned.

💻

Plug In Your GPU

Download OriginLink. Run it in the background. Your GPU starts serving AI inference requests from other users on the network.

GPU is ON → You earn TC
💎

Earn Tensor Credits

Every token your GPU computes earns you Tensor Credits (TC). They accumulate automatically in your account. No manual claims needed.

1 Token Computed = TC Earned
🌙

Use AI When Offline

When your laptop is off, your API key still works. Your requests run on someone else's GPU and are paid for by the Tensor Credits you earned.

GPU is OFF → TC Spent
FLOW DIAGRAM

The Full Picture

Two modes. One account. Both sides of the network explained.

Your laptop is on → OriginLink is running → You are earning

📥
STEP 01Incoming Request

Another user sends an AI request

Someone on the network needs AI—maybe they're writing code, drafting an email, or running a model in their app. They send a request to Origin Of AI.

🔀
STEP 02Smart Routing

The network routes it to your GPU

The Origin Of AI Orchestrator scans active nodes and picks yours based on speed and availability. It sends the job to your machine via OriginLink.

⚙️
STEP 03GPU Compute Running

Your GPU runs the AI model

OriginLink processes the request locally on your GPU. The AI model runs, tokens are generated, and the response streams back to the requester instantly.

💎
STEP 04+ TC Earned

You earn Tensor Credits automatically

For every token your GPU computes, Tensor Credits are deposited into your account in real time. No button to press. No waiting. Automatic.

Result: Your TC balance grows 💎

The more your GPU runs, the more Tensor Credits you stack. Use them to call AI from any device—for free.

THE CREDIT CYCLE

Your Tensor Credit Balance

It goes up when your GPU is on. It goes down when you use the API. Keep contributing more than you consume—and AI stays free.

💻
GPU is On
OriginLink is running, serving requests for others on the network
+ Tensor Credits
💎
Your TC Wallet
120 TC
≈ 100M Input + 100M Output Tokens (8B LLM / 8GB VRAM)
🌙
GPU is Off
Your API calls run on other people's machines — paid by your TC
− Tensor Credits

🔁 The system is self-balancing. The more compute you contribute, the more AI you can consume — effectively for free.

SIMPLE 3-STEP SETUP

Get Started in Minutes

No technical expertise needed. If your machine has a GPU, you're already halfway there.

01
📲

Download OriginLink

Download the OriginLink desktop app for Windows. One installer, one click to connect. Works silently in the background — no configuration required.

02
⚙️

Let Your GPU Work

OriginLink connects your GPU to the network. Whenever it's idle, it serves AI requests for other users. You don't need to do anything — just leave it running.

Runs automatically ✓
03
🚀

Use AI Anywhere

Use your API key to call any supported model from any device. Your Tensor Credits cover the cost. When your GPU is on, credits replenish faster than you spend them.

BEHIND THE SCENES

The Network That Makes It Possible

Origin Of AI Orchestrator maintains a live map of every active OriginLink node worldwide. The moment your request arrives, it's routed to the fastest available GPU in under 10ms.

  • Sub-10ms intelligent node matching by latency & VRAM
  • Encrypted payload — your prompts are never stored to disk
  • 100% OpenAI API compatible — drop-in replacement endpoint
📖API Documentation
P2P MESH ORCHESTRATOR MATRIX● ACTIVE DISCOVERY
Your App / API Key
POST /v1/chat/completions
↓ Request arrives at Origin Of AI ↓
Origin Of AI Orchestrator
Finds best node
↓ Routed to the fastest available GPU ↓
Your GPU Node
Online — earning TC
Another GPU Node
Covers when you're off
↓ Response streams back in < 10ms ↓
💎TC settled instantly
Proof-of-Inference
COMMON QUESTIONS

Straight Answers

Q

Is it really free?

If your GPU earns at least as much TC as your API calls spend, then yes — effectively free. The more you contribute, the more you get. If you use more than you earn, you can top up your TC balance.

Q

What happens when my laptop is off?

Your API key still works. Your requests are served by other active OriginLink nodes on the network, and the cost is deducted from your Tensor Credit balance.

Q

Is my data safe when my GPU serves others?

Yes. All request payloads are encrypted and processed in memory only — nothing is stored to disk on your machine. Your prompts are equally protected when running on others' GPUs.

Q

What GPU do I need?

Any modern Nvidia GPU works — even an RTX 3060 or 4060 with 8GB VRAM can serve lighter models. More VRAM means you can serve larger, higher-paying models.

Q

Do I need to keep OriginLink open all the time?

No. Run it whenever you want — even a few hours a day. The more you run it, the faster your TC balance grows. You're in full control.

Ready to Plug In and Start Earning?

Download OriginLink, connect your GPU, and let it earn Tensor Credits while you sleep. Your AI is waiting.