FreeQwenApi: model tool for Hermes Agent

Local OpenAI-compatible proxy that exposes a Qwen Chat web account as an API for tools such as Hermes Agent

What FreeQwenApi does

FreeQwenApi runs a browser-based session against Qwen Chat and offers a local endpoint that follows the OpenAI chat completions format. You log in through Chromium once, the session is saved under session/, and tools that speak the OpenAI API can then use models such as qwen3.7-max, qwen3.7-plus and qwen3.6-plus. The model list is refreshed from Qwen Chat metadata with npm run models:sync.

It also exposes image generation and video generation endpoints, with video jobs polled through a task status route, plus file upload. Several Qwen accounts can be added, relogged or removed, with OK, WAIT and INVALID statuses and round-robin rotation when limits hit. Health, status and models endpoints and an npm smoke test help check the setup. The README is in Russian and calls this version a fork; it clearly states the project is not a local model and not an official Alibaba or Qwen API, and warns not to commit session files, tokens or .env.

Key features

OpenAI-compatible POST /api/chat/completions on localhost port 3264
Qwen models such as qwen3.7-max and qwen3.7-plus, synced with npm run models:sync
Image and video generation endpoints through Qwen Chat
Multiple accounts with status tracking and round-robin rotation
Config examples for Hermes Agent, LiteLLM, Claude Code and Open WebUI
Health, status and models endpoints plus an npm smoke test

When to use it

Pointing Hermes Agent at Qwen Chat models through a local OpenAI-style endpoint
Connecting Open WebUI or LiteLLM to a Qwen Chat account
Rotating between several Qwen accounts when one hits its limit

Who it is for: Developers who want to use Qwen Chat models from OpenAI-compatible tools, including Hermes Agent, without an official API key.

How it fits with Hermes Agent

The README lists Hermes Agent, LiteLLM and Claude Code as clients with ready-made config examples, so it is a general proxy that supports Hermes.

How to install FreeQwenApi

These commands are copied from the project's README. Check the repository for the latest steps before you run them.

git clone https://github.com/ForgetMeAI/FreeQwenApi
cd FreeQwenApi
npm install
npm run auth

Requirements: Node.js with npm, Chromium for the login step, and a Qwen Chat account.

Note: It works through a logged-in Qwen Chat browser session rather than an official API, and the repository has no license file.

FAQ

What is FreeQwenApi?

FreeQwenApi is a local proxy that turns a Qwen Chat web account into an OpenAI-compatible API endpoint. It is not a local model and not an official Alibaba or Qwen API.

Does FreeQwenApi work with Hermes Agent?

The README lists Hermes Agent among its clients, with ready-made configuration examples for local agents. Hermes can be pointed at the local OpenAI-compatible endpoint.

How do I install FreeQwenApi?

Clone the repository, run npm install, then npm run auth and log in to Qwen Chat in the Chromium window. Run npm run models:sync, then start the server with SKIP_ACCOUNT_MENU=true npm start.