One API key for every frontier model — Nordzhi AI is now live. Explore models →
nordzhiAI
Sign in Get API key
Unified LLM API platform

One API.
Every frontier model.

Access GPT, Claude, Gemini and DeepSeek through a single OpenAI-compatible endpoint. Change your base URL, keep your existing stack, and pay only for the tokens you use.

Get your API key Read the docs
11live models
4providers
1endpoint
USDper-token billing
Live catalog

Frontier models, priced per token

View all models →
gpt-4.1
GPT-4.1
OpenAI
Input / 1M $2.00
Output / 1M $8.00
Context 1,047,576 tokens
Pay as you go
gpt-4.1-mini
GPT-4.1 mini
OpenAI
Input / 1M $0.40
Output / 1M $1.60
Context 1,047,576 tokens
Pay as you go
gpt-4o
GPT-4o
OpenAI
Input / 1M $2.50
Output / 1M $10.00
Context 128,000 tokens
Pay as you go
gpt-4o-mini
GPT-4o mini
OpenAI
Input / 1M $0.15
Output / 1M $0.60
Context 128,000 tokens
Pay as you go
o4-mini
o4-mini
OpenAI
Input / 1M $1.10
Output / 1M $4.40
Context 200,000 tokens
Pay as you go
claude-sonnet-4
Claude Sonnet 4
Anthropic
Input / 1M $3.00
Output / 1M $15.00
Context 200,000 tokens
Pay as you go
claude-3-5-haiku
Claude 3.5 Haiku
Anthropic
Input / 1M $0.80
Output / 1M $4.00
Context 200,000 tokens
Pay as you go
gemini-2.5-pro
Gemini 2.5 Pro
Google
Input / 1M $1.25
Output / 1M $10.00
Context 1,048,576 tokens
Pay as you go
· Why Nordzhi ·

Built for teams that ship

Less integration work, fewer surprises on the bill, and a gateway your existing OpenAI code already understands.

Abstract connected gateway nodes

OpenAI-compatible by design

Point the official OpenAI SDK at https://nordzhi.com/v1 and switch models by changing one string. Tool calls, JSON mode and streaming pass through untouched.

Abstract streaming light ribbons

Streaming without buffering

Server-sent events are proxied as they arrive, so your users see the first token as soon as the upstream model produces it.

Abstract glass tokens and wallet

Token-level, transparent billing

Every request is metered from the prompt and completion tokens reported by the upstream. No seats, no minimums, no expiry on wallet balance.

· Quick start ·

Live in three steps

From sign-up to your first completion in a few minutes — no credit card needed to explore the console.

1

Create your account

Sign up with your email and open the console. Your wallet starts at $0.00.

2

Top up your wallet

Submit a bank-transfer request. Our team replies with account details and credits your balance once the transfer is confirmed.

3

Create a key and call the API

Generate an API key, then call /v1/chat/completions with any model from the catalog.

from openai import OpenAI

client = OpenAI(
    api_key="sk-nordzhi-...",
    base_url="https://nordzhi.com/v1",
)

resp = client.chat.completions.create(
    model="gpt-4.1",
    messages=[{"role": "user", "content": "Hello!"}],
)
print(resp.choices[0].message.content)
curl https://nordzhi.com/v1/chat/completions \
  -H "Authorization: Bearer sk-nordzhi-..." \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-4.1",
    "messages": [{"role": "user", "content": "Hello!"}],
    "stream": true
  }'
import OpenAI from "openai";

const client = new OpenAI({
  apiKey: "sk-nordzhi-...",
  baseURL: "https://nordzhi.com/v1",
});

const resp = await client.chat.completions.create({
  model: "claude-sonnet-4",
  messages: [{ role: "user", content: "Hello!" }],
});
console.log(resp.choices[0].message.content);
OpenAI-compatible · native SSE · token-metered Full API reference →
· Platform ·

Costs you can see. Keys you control.

A console built around prepaid balance and auditable usage — not a monthly subscription you forget about.

Wallet-first billing

Prepaid USD balance, billed per token at each model's published rate. When the balance is empty, requests stop — no surprise invoices.

Live usage logs

Model, prompt and completion tokens, cost, latency and status are recorded for every request and visible in your console.

Keys you control

Create unlimited keys for different apps and environments, then revoke any key instantly from the console.

Human-reviewed top-ups

Bank transfers are checked by our team before credit is applied, keeping the wallet ledger clean and auditable.

11models in the live catalog
4frontier model providers
1OpenAI-compatible endpoint
100%usage-based spending
· FAQ ·

Questions, answered

What is Nordzhi AI?

Nordzhi AI is an AI model API gateway operated by NORDZHI HONG KONG LIMITED. One API key gives you access to GPT, Claude, Gemini and DeepSeek models through the OpenAI-compatible API format.

Is it really OpenAI-compatible?

Yes. The gateway implements /v1/chat/completions and /v1/models with the same request and response schema, including server-sent event streaming. If your code works with OpenAI, change the base URL and key.

How does billing work?

Your wallet is prepaid in USD. Each request is billed per token at the model's published rate, using the usage figures returned by the upstream provider. Prompt and completion tokens are metered separately, and every charge appears in your usage log.

How do I top up my wallet?

Open the wallet in your console and submit a recharge request. Our team replies by email with bank account details; your balance is credited once the transfer is confirmed. Wallet balances do not expire.

Which models can I call?

See the Models page for the live catalog. Model availability and per-token rates are managed in the admin console, so the list can change without any code changes on your side.

What happens if a model is unavailable?

If no upstream channel is currently serving a model, the API returns a clear 503 error rather than a silent failure. Contact support and we will help route the request to an available model.

· Start building ·

Ship your next AI feature today

One endpoint. Every model. Usage-based pricing that starts from zero.

Create free account See pricing