Sophomoresty

Convert Google Gemini web into OpenAI-compatible API. Zero auth, cross-platform, single file.

200
93
69% credibility
Found May 28, 2026 at 200 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

This project creates a local bridge that lets you use Google's free Gemini AI through any app designed for OpenAI's API, enabling free AI access with existing tools.

How It Works

1
🔍 You discover a free AI option

You hear about using Google's free Gemini AI with your existing apps that normally need paid API access.

2
📥 You download and start the tool

You run a simple command to launch a local server on your computer that acts as a bridge.

3
🔌 You connect your favorite AI app

You enter your local server address into any app designed for AI assistants, like chat tools or coding programs.

4
You choose your access level
🚀
Quick start (free)

Use all the fast AI models right away without any setup

🔑
Full access (optional)

Add your Google account cookies to unlock the most powerful AI models

5
💬 You chat with AI as usual

Your app sends messages to your local server, which quietly connects to Google's free AI service.

You get free AI responses

You receive thoughtful answers from Google's AI through your familiar apps, without paying anything or creating accounts.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 200 to 200 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is gemini-web2api?

gemini-web2api is a Python proxy server that converts Google Gemini's web interface into an OpenAI-compatible API. Drop it on any machine, point your existing OpenAI client at it, and you get access to Gemini's models without paying for API keys. It speaks the `/v1/chat/completions` protocol natively, so tools like Cherry Studio, ChatBox, or the OpenAI Python SDK work out of the box. You can switch models with a simple config change, enable streaming, and even use function calling. A free Google account is all you need.

Why is it gaining traction?

The hook is simple: completely free access to Gemini's models, including the thinking variants that produce 20k+ character responses. Developers are tired of API billing surprises, and this sidesteps that entirely by routing through Gemini's public web endpoint. The setup takes one command: `python gemini_web2api.py`. No Docker required, no API keys to manage, no paid subscription. It also supports Gemini CLI and OpenAI Codex CLI integration, making it a flexible backend for different workflows.

Who should use this?

Developers prototyping AI features who want to avoid API costs will get the most value. Teams already locked into OpenAI client libraries can migrate to Gemini without rewriting integration code. Researchers experimenting with Gemini's deep thinking modes will appreciate the zero-friction access. It's less suited for production systems where stability matters more than cost.

Verdict

gemini-web2api fills a real niche for free, low-friction Gemini access, but the 200-star count signals a project still finding its footing. The credibility score of 0.699999988079071% reflects this early-stage status. Documentation is clear and the stdlib-only design is elegant, but relying on reverse-engineered web endpoints means updates to Gemini's site could break things overnight. Use it for experimentation and prototyping. For anything mission-critical, wait for a more battle-tested release or consider Google's official API.

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.