The last revolution was metered in kilowatt-hours.
This one is metered in tokens.
AI Power Grid meters generative compute — text, images, video, and workflows drawn from community GPUs, with worker payouts settled on Base. Core routing is coordinated today; validator evidence and partner-run nodes are the path beyond that control plane.
Build on The Grid
Agents, automations, art, video, workflows — running on community GPUs today. Pick one and ship.
Generate Images
Create AI art with z-image-turbo, FLUX.2 Klein, and Krea 2 Turbo. Curated open image models.
aipg.art →Run Agents & Workflows
Open-source LLMs powering your agents, automations, and AI workflows. gpt-oss, Qwen3, DeepSeek — no corporate worldview baked in.
aipg.chat →Build with the API
OpenAI and Anthropic compatible text plus Grid image, video, and audio endpoints. Community-powered backend with daily and prepaid credits.
API Docs →Earn with your GPU
Run an LLM, image, or video worker. Earn from the current hourly AIPG payout rail. No bond required to start; future bonding policy is not finalized.
Run a Worker →Text, image, and video generation through familiar APIs. Three lines to get started:
For onchain builders
The AI generation layer crypto is missing.
Onchain apps need generative AI for agents, games, wallets, NFT tools, social apps, creator markets, and autonomous services. Today they either call a closed AI vendor, run their own GPU stack, or stitch together fragile one-off providers.
AI Power Grid gives them a crypto-native alternative: familiar generation APIs, distributed GPU supply, transparent worker payments, and a path toward validator-backed quality.
Text, image, video, workflows
One network for chat, agents, image generation, video generation, and reproducible creative pipelines.
Familiar APIs at the edge
Builders use OpenAI-compatible endpoints and the SDKs they already know, without becoming GPU operators.
Crypto-native underneath
Base settlement, worker rewards, model and workflow registries, validator evidence, and a path toward trustless claims.
AI shouldn't have a doorman.
$20 a month before your first line of code. A trial that expires. A card you don't have. Closed AI is full of velvet ropes. AIPG is built so the people who can pay fund the people who can't — and so builders can choose open APIs backed by distributed GPU supply instead of one closed vendor.
Start without a subscription
Eligible accounts can use a daily allowance, then add prepaid credits when they need more capacity. The current terms are visible in the console before a request is sent.
Paid usage sustains access
Paid credits fund network operations and help us keep a practical daily allowance available while independent workers are compensated for completed work.
Open models. Honest boundaries.
Build with open models without tying your application to one closed provider. Confidential GPU sessions are planned; current community workers are not a private-inference guarantee.
Plug in your GPU. Power the Grid. Earn.
The network runs on real hardware that real people plug in. Two ways to take part: run a Generator to contribute compute and earn, or a Validator to independently test it. The text worker is live today; the managed media installer stays gated until signed hardware profiles qualify.
Generator
Connect an existing text backend to the current AIPG worker payout rail. Experienced media operators can run reviewed workflows, while managed media setup remains gated.
Validator
Run independent, Grid-assigned checks against worker outputs and sign the evidence. The preview is CPU-only and intentionally has no routing, payout, strike, slash, or reward authority.
Got questions? Join our Discord and talk to the community.
How The Grid Works
Community GPUs run AI models. Smart routing sends your request to an available worker. Workers currently share an hourly AIPG budget according to recorded work.
GPU Workers
Gaming rigs to data centers can join when their OS, backend, model, and available memory pass the worker's local checks.
Smart Routing
Requests automatically route to the best available worker based on model, load, and latency.
Token Rewards
Workers earn from an hourly AIPG budget according to recorded generative work. Multi-asset payouts are planned, not live.
Toward distributed coordination
In DevelopmentEvery centralized AI you've used can be unplugged by a company, a regulator, or a Stripe ban. AIPG is moving beyond its current coordinated Core through partner-run nodes, verifiable evidence, and progressively on-chain settlement and control.
- ◆On-chain job anchoring with cryptographic proofs
- ◆Validator nodes submit assignment-bound evidence; bonding and rewards come later
- ◆Planned worker-claimed rewards through audited smart contracts
- ◆P2P network with no single point of failure
Privacy at the hardware level
Coming with BlackwellYour medical notes, your legal docs, your agent's API keys — they shouldn't be readable by the stranger running the GPU. NVIDIA Confidential Computing on Blackwell (B200/B300) encrypts your workload in GPU memory and proves cryptographically it ran in a secure enclave.
- ◆Hardware-isolated Trusted Execution Environment (TEE)
- ◆GPU memory encrypted end-to-end during inference
- ◆Cryptographic attestation proves genuine secure enclave
- ◆Run sensitive workloads on untrusted infrastructure
Open Source, Open Doors
Public contracts, SDKs, worker and validator clients, models, and workflow specifications are open to inspect and build on.
Published interfaces use familiar APIs, so builders can integrate without tying their application to a proprietary client.
No Vendor Lock-In
Switch providers anytime. Run your own instance. Your data, your choice.
Faster Innovation
Every contribution benefits the whole network. Improvements compound across all users.
Community Operated
Independent operators can run workers and preview validators. Core coordination is still moving toward trusted partner nodes.
Global Access
Same API whether you're in Buenos Aires or Bangalore. No waitlists, no geographic restrictions, no approval process.
Earn with Your GPU
GPU owners earn AIPG today for running text, image, and video jobs. Real work, publicly verifiable payouts. Your hardware serves actual users instead of sitting idle.
Useful Work
Recorded completed jobs contribute to each worker's reward share. GPUs do productive work instead of arbitrary calculations.
Fair Launch
AI Power Grid launched with no VC, no presale, no pre-mine. Community-first from day one.
Dec 2023Building the Grid
Launched aipg.art for image generation, aipg.chat for LLM conversations, and the first GPU worker network. Public beta shaped by community feedback.
2024Base L2 Migration
Migrated to Base L2 with 150M fixed-supply AIPG token. Deployed StakingVault and core infrastructure contracts. The passive staking program later ended; withdrawals remain available.
Q1–Q2 2025Grid SDK & Dashboard
Released the Grid SDK for developers. Developer console and API explorer went live at console.aipowergrid.io. On-chain model registry for worker discovery.
Q3–Q4 2025Streaming API
✓ Live NowReal-time token streaming with OpenAI and Anthropic compatibility. Familiar SDK integration, fast image generation with z-image-turbo, and text-to-video via LTX-2.3.
Q1 2026Validator Evidence & Worker Trust
Grid-issued text assignments, targeted probes, scorecards, and a non-economic quorum lifecycle. P2P dispatch and worker bonding remain gated prototypes, not production claims.
Q2 2026Validator Distribution & Partner Nodes
Package an easy validator release, prove independent multi-validator operation, and harden the path toward trusted partner-run core nodes.
Q3 2026On-Chain Verification & NFTs
Worker registry and job escrow on Base L2. AI art NFTs with full generation parameters on-chain — not just a JPEG, the actual recipe. Dispute resolution via stake slashing.
Q4 2026Confidential AI
Private AI sessions using trusted execution environments. Verifiable, private inference where operators can't see your prompts or outputs.
2027