OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI APIs & Platforms

Webai2Api

WebAI2API is a JavaScript application that uses Camoufox and Playwright to interact with web-based AI services and expose supported functions through OpenAI-compatible API endpoints. It includes a visual management interface and supports concurrent browser windows with isolated account data.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Webai2Api — OpenViST project profile in AI APIs & Platforms.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when a project needs an OpenAI-compatible interface to supported AI services accessed through their websites, and you can operate browser sessions and complete the services' account login steps.

Who it is for

  • Developers connecting applications to supported web-based AI services through OpenAI-format requests.
  • Operators who can self-host a browser-automation service and manage website logins and sessions.

Documented uses

  • Send text-generation requests to configured web AI services using an OpenAI-compatible chat endpoint.
  • Discover currently configured models through the models endpoint.
  • Submit documented image inputs through the chat interface where supported by the connected service.
  • Manage browser adapters, logs and isolated account sessions through the WebUI.
Before you choose

Requirements & limits

The project documents source and Docker deployment. Initial use requires manually signing in to the relevant AI website account. The service API and WebUI use the configured authentication token.

  • The application automates website interactions rather than calling the AI providers' native APIs.
  • The documented source environment requires Node.js v20 or later; Linux setups may require additional display and browser libraries.
  • Streaming responses include heartbeat support, while queued non-streaming requests can be rejected when the queue is overloaded.
  • Connected services differ in text, image and video capabilities; the current configured model list is available at GET /v1/models.
  • The WebUI transport is unencrypted, so remote public access should be protected with HTTPS or an SSH tunnel.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Set up WebAI2API from source

Install and start the service, then prepare its management interface for a connected AI website account.

Before you start

  • Node.js v20.0.0 or later
  • Windows, Linux or macOS
  • pnpm and network access to download browser resources during initialization
  • On Linux, the listed Xvfb, VNC and browser libraries

Steps

  1. Install dependencies with pnpm install.

  2. Run the command below to obtain browser resources; the project documents an optional proxy argument for this step.

    Documented command
    npm run init

  3. Start the service with the command below.

    Documented command
    npm start

  4. Configure the API token in data/config.yaml, restart the program, open the WebUI, and manually sign in to the required AI website account.

Expected outcome

The service is started and the connected website account is initialized for subsequent API use.

Guide limitations

  • Configuration changes require restarting the program.
  • Linux and Docker users may need to connect to the virtual display to log in.
  • The documentation says to complete any required human verification and initial site prompts manually.
  • Public WebUI access should use HTTPS or an SSH tunnel because its transport is unencrypted.
Documentation 1 ↗
Source-based practical guide

Send a streaming chat request

Submit a text conversation to a configured service through the OpenAI-compatible chat endpoint.

Before you start

  • A running WebAI2API service
  • A configured API authentication token
  • A model name available from GET /v1/models

Steps

  1. Retrieve the configured model names from GET /v1/models.

  2. Send a JSON request to POST /v1/chat/completions with the chosen model, a messages array, and stream: true.

  3. Include the configured token as a Bearer authorization value.

Expected outcome

The endpoint returns a streaming chat response with heartbeat support.

Guide limitations

  • Processing time depends on browser interactions and can vary.
  • The documentation recommends streaming; queued non-streaming requests may be rejected when the configured queue limit is exceeded.
  • Available models and capabilities depend on the current configuration and connected website.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
16 October 2026
Original website title
foxhui/WebAI2API · Official homepage title · retained in its source language
How we review sourcesSuggest a correction