One key. Every LLM. Prepaid credits.

The simplest way to ship
AI in production.

Point your existing OpenAI SDK at our gateway and get instant access to GPT, Claude, and Gemini. One bill, transparent pricing, analytics included. No provider accounts to manage.

Get started See pricing

Setup in 60 seconds. Top up when you are ready to ship.

Three lines. That's it.

Change baseURL in your existing OpenAI client. We handle routing, billing, and analytics.

app.ts

import OpenAI from 'openai';
import { ScopeVeil } from '@scopeveil/sdk';

const ai = new OpenAI({
  apiKey: process.env.SCOPEVEIL_KEY,
  baseURL: ScopeVeil.gatewayBaseURL(),
});

const r = await ai.chat.completions.create({
  model: 'anthropic/claude-sonnet-4-6',
  messages: [{ role: 'user', content: 'hi' }],
});

Response

One bill. Every model.

Stop juggling OpenAI, Anthropic, and Google billing dashboards. Top up once, pay one provider, route per request.

Privacy by default.

Prompt content never gets logged. Only metadata flows to analytics. Open-source SDK alternative if you'd rather not proxy.

Analytics included.

Cost-per-feature, latency percentiles, error rates per model. Same dashboard whether you use the gateway or the SDK.

Pricing that fits in one sentence.

Pay what the provider charges, plus a small markup. No subscription, no monthly fees.

Coming from another gateway?

We wrote honest, side-by-side comparisons against the gateways and patterns most teams evaluate. No marketing fluff. Where the alternative wins, we say so.

vsOpenRouter→vsHelicone→vsLiteLLM→vsOpenAI direct→

Ship your first request in 60 seconds.

Get started

The simplest way to ship AI in production.