Labol Tech Docs

Labol Tech is a decentralized AI compute network on Solana. Miners share their AI API capacity to serve inference requests. Users burn $LABOL tokens to access AI inference — no subscriptions, no credit cards, no API keys needed.

How It Works

The protocol has two sides:

• Miners — install the relay client, connect their Solana wallet + AI API key, and serve inference requests routed by the coordinator. Your API key never leaves your machine.
• Users — burn $LABOL tokens to receive compute credits. Use those credits to call the OpenAI-compatible API endpoint and receive AI responses.

Architecture

[Miners]  ──  WebSocket  ──►  [Coordinator]
                                     │
[Users]   ──  burn $LABOL  ──►       │  ──►  [Relay Router]  ──►  [AI Response]
                                     │
                                [Supabase DB]

Quick Start

Get started as a miner or user in under 5 minutes.

As a Miner

# Install
npm install -g labol-relay

# Start mining
labol-relay start \
  --wallet YOUR_SOLANA_WALLET \
  --ai-key  sk-YOUR_API_KEY \
  --ai-url  https://api.openai.com/v1 \
  --model   gpt-4o-mini

As a User

How Mining Works

Labol miners don't need GPUs. They need API keys.

There are two ways to earn $LABOL:

• Comprehension Mining — solve multi-hop reasoning challenges generated by the coordinator. No AI key needed. Earn base rewards per correct submission.
• Relay Mining — serve AI inference requests from users. Requires an AI API key. Earn 2x reward weight per fulfilled request.

Supported Providers

Relay Setup

labol-relay start \
  --wallet      YOUR_SOLANA_WALLET \
  --ai-key      sk-YOUR_API_KEY \
  --ai-url      https://api.openai.com/v1 \
  --model       gpt-4o-mini \
  --coordinator wss://api.laboltech.xyz

The relay client connects to the coordinator via WebSocket, registers your wallet + model, and begins receiving inference requests. Your API key is used locally — it never leaves your machine.

Burn $LABOL

Burn tokens to get compute credits. 1 $LABOL = 1,000 output tokens of AI inference.

Check your balance:

curl "https://api.laboltech.xyz/v1/burn/balance?wallet=YOUR_WALLET"

// Response:
{
  "wallet": "YOUR_WALLET",
  "compute_credits": 50000,
  "tokens_burned": 50
}

API Keys

Generate a lk_xxx key to authenticate with the compute API.

curl -X POST https://api.laboltech.xyz/v1/apikeys/create \
  -H "Content-Type: application/json" \
  -d '{"wallet":"YOUR_WALLET"}'

// Response:
{
  "key": "lk_...",
  "wallet": "YOUR_WALLET",
  "token_balance": 0
}

Or use the Dashboard to generate and manage keys visually.

Compute API

OpenAI-compatible. Change the base URL. Zero code rewrites.

Python

from openai import OpenAI

client = OpenAI(
    base_url = "https://api.laboltech.xyz/v1",
    api_key  = "lk_YOUR_KEY",
)

response = client.chat.completions.create(
    model    = "auto",
    messages = [{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)

cURL

curl -X POST https://api.laboltech.xyz/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "X-Api-Key: lk_YOUR_KEY" \
  -d '{
    "model": "auto",
    "messages": [{"role": "user", "content": "Hello"}],
    "max_tokens": 512
  }'

API Reference

Base URL

https://api.laboltech.xyz

Auth

Mining

Compute

Burn

API Keys

Tokenomics

Contracts