Scale your AI apps with instant credit billing
The easiest way to meter and charge for LLM usage. No infrastructure needed. Drop in our SDK and start monetizing immediately.
import { NexusGateway } from '@nexus/server';
// Initialize the gateway
const gateway = new NexusGateway(process.env.NEXUS_KEY);
async function generateResponse(req) {
// Check user balance & reserve credits
const session = await gateway.metering.start({
userId: req.user.id,
model: 'gpt-4-turbo'
});
try {
const stream = await llm.generate(req.prompt);
// Stream and bill simultaneously
return session.streamWithBilling(stream);
} catch (err) {
await session.refund();
}
}
Engineered for Modern AI
Everything you need to handle complex LLM economics.
Real-time Metering
Calculate token costs dynamically as they stream. Ensure you never over-provision or under-charge, down to the exact fraction of a cent.
Managed Accounts for Your Customers
Meter, govern, and protect the customer accounts you operate from one workspace while keeping balances, limits, and usage attribution in sync.
Privacy by Design
GDPR-aligned data handling, encrypted storage for sensitive provider credentials, and audit-ready controls protect the information your operation relies on.
Simple, Transparent Pricing
Pay only for the throughput you secure.
Starter
For hobbyists and early prototypes.
- check Up to 10k requests/mo
- check Standard models
- check Community support
Pro
For growing applications in production.
- check Up to 1M requests/mo
- check Advanced model routing
- check Stripe integration
- check Priority email support
Enterprise
Custom limits for massive scale.
- check Unlimited requests
- check Custom integrations
- check SLA guarantees
- check Dedicated success manager