ChatGPT API Portal

A local orchestration proxy that provisions private browser workers, keeps ChatGPT sessions hot, and exposes unified OpenAI-compatible API routes.

Create account Sign in

Try it now

No account needed

Streaming

Send a message straight to ChatGPT through a real browser. Answers stream back token by token. ChatGPT sometimes restricts logged-out use - sign in to connect your own account when it does.

Ask anything to get started.

Sign In

Enter credentials to unlock the manager control panel

Don't have an account? Register now

Create Account

All portal credentials must be completed below

ProtonVPN Connection Parameters

Already registered? Sign in here

Orchestration Dashboard

Worker Status

Loading...

Retrieving properties...

ChatGPT Session

Not logged in

Orchestration Console Log

Test Playground

Streamed Response

No response yet.

Account Settings

Profile Information

Portal Password

API Credentials

Use this key to authorize API integrations. Keep it secure and do not share it.

Key Preview: Loading...

Key created: Loading...

ProtonVPN Credentials

Active Workers

Name Email Status Port Last active
Loading active workers...

API Documentation

Authentication

All requests must supply your API key as a bearer authorization token: 

Authorization: Bearer YOUR_API_KEY

API Endpoints

POST /v1/responses POST /v1/responses

The OpenAI Responses API. Point any OpenAI SDK at /v1 and it works unchanged. Supports streaming, previous_response_id for multi-turn, function tools, the built-in {"type":"web_search"} tool, and text.format for structured output. Companion routes: GET/DELETE /v1/responses/{id}, POST /v1/responses/{id}/cancel, GET /v1/models, GET /v1/conversations, and POST /v1/personalization/disable.

curl -X POST "https://YOUR_HOST/v1/responses"   -H "Authorization: Bearer YOUR_API_KEY"   -H "Content-Type: application/json"   -d '{"model":"gpt-4o","input":"Say hello"}'

POST /v1/messages POST /v1/messages

Send a prompt directly into your running ChatGPT browser session.

Field Description
message Required. The text message content to send.
chatId Optional. Continuation chat ID to keep dialog thread context.
searchWeb Optional. Boolean parameter to activate Web Search.
Click to view cURL snippet 
curl -X POST "http://localhost:3000/v1/messages" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d "{\"message\":\"Hello\",\"searchWeb\":false}"

POST /v1/chat/completions POST /v1/chat/completions

Send messages in standard OpenAI API format compatibility.

Field Description
messages Required array. Standard OpenAI array of chat messages.
response_format Optional. Structured output settings.
Click to view cURL snippet 
curl -X POST "http://localhost:3000/v1/chat/completions" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d "{\"messages\":[{\"role\":\"user\",\"content\":\"Hello\"}]}"

GET /v1/account

Retrieve the account profile information mapping the Bearer credentials key.

Response Codes

Status Details
400 Bad Request validation_error
401 Unauthorized invalid_credentials or session_expired
502 Bad Gateway worker_unreachable (worker server down or loading)
500 Server Error internal_server_error