CoreAI / Model-Derived Infrastructure

One model layer.
Every workflow you need.

Route text and reasoning models through one reliable OpenAI-compatible gateway. Track model choice, usage, and cost in one place — pay only for the tokens you use.

Get started View models & pricing
11+frontier models
<80msgateway overhead
100%usage-based billing
Model gateway / demo 200 OK
01 /Client
Your app
sk-coreai-••••••••
02 /Gateway
CoreAI
coreai-balanced
03 /Model
Provider
gpt-4.1OpenAI
Route
Why CoreAI

The model layer for every production workflow.

01 / Drop-in replacement

Point your SDK at one URL

Use https://coreai.win/v1 as the base URL in any OpenAI-compatible client and switch models by changing one string. Streaming, tool calls and JSON mode pass through untouched.

02 / Native streaming

First token, as fast as upstream

Server-sent events are proxied byte-for-byte with zero buffering, so the first token arrives as fast as the provider can send it.

03 / Token-level billing

Metered on reported usage

Usage is metered on the exact prompt_tokens / completion_tokens returned by the upstream — nothing estimated, everything visible in live logs.

04 / Keys you control

Unlimited keys, instant revoke

Create separate keys per app or agent, rotate or delete them any time. Every request is authenticated and scoped to your wallet balance.

CoreAI / Model Catalog

Select a model

Compare model families, supported APIs, and per-token pricing across the gateway.

11 MODELS LIVE
01 / Models
02 / Selected model · OpenAI

gpt-4.1

GPT-4.1 — use one API key with the supported endpoints below, billed per token from your prepaid wallet.

03 / API Protocols
POST/v1/chat/completionsChat Completions
GET/v1/modelsList models
04 / Token rates
Input per 1M prompt tokens $2.00
Output per 1M completion tokens $8.00
Context token window 1,047,576
Model and pricing data update in real time. Create an API key Full pricing →
02 / Selected model · OpenAI

gpt-4.1-mini

GPT-4.1 mini — use one API key with the supported endpoints below, billed per token from your prepaid wallet.

03 / API Protocols
POST/v1/chat/completionsChat Completions
GET/v1/modelsList models
04 / Token rates
Input per 1M prompt tokens $0.40
Output per 1M completion tokens $1.60
Context token window 1,047,576
Model and pricing data update in real time. Create an API key Full pricing →
02 / Selected model · OpenAI

gpt-4o

GPT-4o — use one API key with the supported endpoints below, billed per token from your prepaid wallet.

03 / API Protocols
POST/v1/chat/completionsChat Completions
GET/v1/modelsList models
04 / Token rates
Input per 1M prompt tokens $2.50
Output per 1M completion tokens $10.00
Context token window 128,000
Model and pricing data update in real time. Create an API key Full pricing →
02 / Selected model · OpenAI

gpt-4o-mini

GPT-4o mini — use one API key with the supported endpoints below, billed per token from your prepaid wallet.

03 / API Protocols
POST/v1/chat/completionsChat Completions
GET/v1/modelsList models
04 / Token rates
Input per 1M prompt tokens $0.15
Output per 1M completion tokens $0.60
Context token window 128,000
Model and pricing data update in real time. Create an API key Full pricing →
02 / Selected model · OpenAI

o4-mini

o4-mini — use one API key with the supported endpoints below, billed per token from your prepaid wallet.

03 / API Protocols
POST/v1/chat/completionsChat Completions
GET/v1/modelsList models
04 / Token rates
Input per 1M prompt tokens $1.10
Output per 1M completion tokens $4.40
Context token window 200,000
Model and pricing data update in real time. Create an API key Full pricing →
02 / Selected model · Anthropic

claude-sonnet-4

Claude Sonnet 4 — use one API key with the supported endpoints below, billed per token from your prepaid wallet.

03 / API Protocols
POST/v1/chat/completionsChat Completions
GET/v1/modelsList models
04 / Token rates
Input per 1M prompt tokens $3.00
Output per 1M completion tokens $15.00
Context token window 200,000
Model and pricing data update in real time. Create an API key Full pricing →
02 / Selected model · Anthropic

claude-3-5-haiku

Claude 3.5 Haiku — use one API key with the supported endpoints below, billed per token from your prepaid wallet.

03 / API Protocols
POST/v1/chat/completionsChat Completions
GET/v1/modelsList models
04 / Token rates
Input per 1M prompt tokens $0.80
Output per 1M completion tokens $4.00
Context token window 200,000
Model and pricing data update in real time. Create an API key Full pricing →
02 / Selected model · Google

gemini-2.5-pro

Gemini 2.5 Pro — use one API key with the supported endpoints below, billed per token from your prepaid wallet.

03 / API Protocols
POST/v1/chat/completionsChat Completions
GET/v1/modelsList models
04 / Token rates
Input per 1M prompt tokens $1.25
Output per 1M completion tokens $10.00
Context token window 1,048,576
Model and pricing data update in real time. Create an API key Full pricing →
02 / Selected model · Google

gemini-2.5-flash

Gemini 2.5 Flash — use one API key with the supported endpoints below, billed per token from your prepaid wallet.

03 / API Protocols
POST/v1/chat/completionsChat Completions
GET/v1/modelsList models
04 / Token rates
Input per 1M prompt tokens $0.30
Output per 1M completion tokens $2.50
Context token window 1,048,576
Model and pricing data update in real time. Create an API key Full pricing →
02 / Selected model · DeepSeek

deepseek-chat

DeepSeek V3 — use one API key with the supported endpoints below, billed per token from your prepaid wallet.

03 / API Protocols
POST/v1/chat/completionsChat Completions
GET/v1/modelsList models
04 / Token rates
Input per 1M prompt tokens $0.27
Output per 1M completion tokens $1.10
Context token window 64,000
Model and pricing data update in real time. Create an API key Full pricing →
02 / Selected model · DeepSeek

deepseek-reasoner

DeepSeek R1 — use one API key with the supported endpoints below, billed per token from your prepaid wallet.

03 / API Protocols
POST/v1/chat/completionsChat Completions
GET/v1/modelsList models
04 / Token rates
Input per 1M prompt tokens $0.55
Output per 1M completion tokens $2.19
Context token window 64,000
Model and pricing data update in real time. Create an API key Full pricing →
OpenAI-compatible setup

Make your first model call in three steps.

Create an API key, set the base URL, and send your first request.

01

Create your API key

Create a separate key for each app, tool, or agent from the console.

sk-coreai-••••••••
02

Set the base URL

Point your existing OpenAI-compatible client at the gateway.

https://coreai.win/v1
03

Send your first request

Use any model ID from the catalog as the model field.

POST /v1/chat/completions
quickstart.sh
curl https://coreai.win/v1/chat/completions \
  -H "Authorization: Bearer sk-coreai-••••••••" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-4.1",
    "messages": [{"role":"user","content":"Hello!"}]
  }'
from openai import OpenAI

client = OpenAI(
    api_key="sk-coreai-••••••••",
    base_url="https://coreai.win/v1",
)

resp = client.chat.completions.create(
    model="gpt-4.1",
    messages=[{"role": "user", "content": "Hello!"}],
)
print(resp.choices[0].message.content)
import OpenAI from "openai";

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

const resp = await client.chat.completions.create({
  model: "gpt-4.1",
  messages: [{ role: "user", content: "Hello!" }],
});
console.log(resp.choices[0].message.content);
Usage / Cost Visibility

Usage and costs at a glance.

See API keys, request logs, usage, and spend in one console — with a prepaid wallet that never expires.

01 / Keys

Separate keys for clearer access.

Manage access separately for apps, tools, and agents. Create unlimited keys, rotate or revoke instantly.

02 / Logs

Detailed records for every request.

Status, latency, token usage, and cost for every request — metered on provider-reported usage, visible in real time.

03 / Wallet

Prepaid balance, no surprises.

Recharge by bank transfer, watch requests bill down to four decimal places, and top up again whenever you need.

Ready for commercial use

Ship your next AI feature today

One endpoint. Every model. Usage-based pricing that scales down to zero.

Create free account
FAQ

Questions, answered

Which models are supported?

GPT-4.1, GPT-4o, o4-mini, Claude Sonnet 4, Claude 3.5 Haiku, Gemini 2.5 Pro / Flash and DeepSeek V3 / R1 are available at launch, with more added regularly. Check the models page for the live catalog.

How does billing work?

Prepaid wallet, billed per token at each model's published rate. Prompt and completion tokens are metered separately using the usage reported by the upstream provider — nothing is estimated.

How do I recharge my wallet?

Submit a recharge request in the console and leave your contact email. Our support team replies with bank account details; once your transfer is confirmed your balance is credited. See the docs for details.

Is the API OpenAI-compatible?

Yes. The gateway implements /v1/chat/completions and /v1/models with the same request and response schema, including streaming via server-sent events.

Can I rotate or revoke keys?

Absolutely. Create unlimited keys from the console, and delete a key at any time to revoke it instantly.