One API for
every model

A paid AI reverse-proxy that speaks both the OpenAI and Anthropic protocols — streaming or not — behind a single base URL with transparent, token-based pricing.

Get your API key View docs

gpt-4ogpt-4.1claude-sonnet-4claude-opus-4o3gemini-2.5-prollama-3.3mistral-largegpt-4ogpt-4.1claude-sonnet-4claude-opus-4o3gemini-2.5-prollama-3.3mistral-large

Scroll

The proxy layer

Everything routes through one endpoint

Swap providers without touching your code. Kurol Proxy normalizes every model behind a single, compatible surface.

One base URL

Point your SDK at api.kurolproxy.com/v1 and reach every provider. We handle routing, translation and fan-out.

OpenAI & Anthropic compatible

Drop-in for both protocols. Keep your existing OpenAI or Anthropic SDK and just change the base URL.

Streaming + non-streaming

Server-sent token streaming or buffered JSON responses — your call, on every supported model.

Token-based pricing

Pay for the tokens you use. No seats, no surprises — usage you can actually reason about.

Per-key usage & limits

Mint scoped keys with their own daily token limits and watch consumption in real time.

Discord login

One-click auth with Discord. No passwords to manage — sign in and your dashboard is ready.

Quickstart

Three lines to switch

Keep the SDK you already use. Set your key, change the base URL, and call any model by name.

https://api.kurolproxy.com/v1

Get your API key

app.ts

1import OpenAI from "openai";
2
3const client = new OpenAI({
4  apiKey: process.env.KUROL_API_KEY,
5  baseURL: "https://api.kurolproxy.com/v1",
6});
7
8const res = await client.chat.completions.create({
9  model: "claude-sonnet-4",
10  messages: [{ role: "user", content: "Hello, Kurol" }],
11  stream: true,
12});

Pricing

Plans for every workload

Token-based, no surprises. Billing is handled in our Discord via a quick ticket.

Coming soon

Plans are on the way

We're finalizing our tiers. In the meantime, sign in with Discord to get started, or open a ticket for access.