AI Inference for
Everyone
Run open-source LLMs at 1/5 the cost of OpenAI. Powered by a decentralized network of GPU providers earning crypto rewards.
Also available for macOS · All platforms
Network Status
Real-time stats from the Cynthetic network
How It Works
Three simple steps. No infrastructure to manage.
Call the API
Send inference requests using the same OpenAI-compatible format you already know.
Platform Routes
Our scheduler picks the fastest eligible node based on reputation, stake, and latency.
Node Executes
A GPU node runs the model locally and returns the result. You pay AIC; the node earns AIC.
Built for Developers
If you've used the OpenAI API, you already know how to use Cynthetic.
import OpenAI from 'openai';
const client = new OpenAI({
baseURL: 'https://api.cynthetic.ai/v1',
apiKey: 'aic_your_key_here',
});
const response = await client.chat.completions.create({
model: 'llama3-8b',
messages: [{ role: 'user', content: 'Hello!' }],
});
console.log(response.choices[0].message.content);Zero Migration
Swap your base URL — everything else stays the same. Works with LangChain, LlamaIndex, and any OpenAI SDK.
Pay Per Token
No monthly minimums. Start free with 60 AIC (~50K tokens/week). Scale when you need to.
Multi-Model Access
Llama 3, Mistral, Qwen, Gemma — all through one API. Switch models with a parameter change.
Turn Your GPU into a Money Printer
Download the desktop app. Click start. Earn crypto while your PC is idle.
Download the Agent
One-click install for Windows, Mac, and Linux.
Auto GPU Detection
Detects your hardware and recommends the best model for your specs.
Go Online
Connect to the network and start accepting inference tasks.
Earn AIC
Get paid in AIC tokens for every task your node completes.
Estimated Earnings
* Earnings depend on network demand, model size, and uptime. Early operators get 3x bonus.
Minimum Requirements
Simple, Transparent Pricing
Start free. Scale as you grow. Always cheaper than centralized alternatives.
Free
Perfect for experimenting and prototyping.
- 60 AIC/week (~50K tokens)
- All open-source models
- OpenAI-compatible API
- Community support
- 120 req/min rate limit
Pro
For developers shipping AI-powered products.
- 5,000 AIC/month (~4M tokens)
- All models + priority routing
- Higher rate limits (600 req/min)
- Usage analytics dashboard
- Email support
- No cold-start delays
Enterprise
For teams needing SLA guarantees and scale.
- Unlimited tokens
- 99.9% uptime SLA
- Dedicated node pool
- Custom models & fine-tunes
- Dedicated account manager
- SOC 2 compliance (roadmap)
Ready to Build with Decentralized AI?
Join hundreds of developers saving 80% on inference costs, or start earning with your GPU today.
No credit card required. Start in under 2 minutes.