ZenithBar is a premium, all-in-one desktop companion for power AI users. It combines a sophisticated AI Quota Monitor (Zenith) with a high-performance Local AI Gateway (Zenith Nexus).
Designed for developers and enthusiasts who rely on multiple AI services, ZenithBar provides a unified interface to monitor your limits, balance request loads, and maintain absolute privacy over AI interactions.
| Main Dashboard (Light) | Nexus Providers |
|---|---|
 |
 |
| Nexus Playground | Connection Details |
|---|---|
 |
 |
| Settings (About) | Dashboard (Dark) |
|---|---|
 |
 |
ZenithBar represents a personal turning point in my career.
After years of working as a private developer, I've decided to step out of the shadows and share my first open-source project. For a long time, I didn't dare to do this, thinking my knowledge "wasn't good enough" or that I wasn't ready to participate in the open-source community.
I've realized that the best way to change that is by doing. ZenithBar is my transition to a more collaborative way of building. My goal with this project is to create a space for mutual learning. I'm not here to show I know everything; I'm here to learn alongside anyone who wants to contribute.
If this project grows, I want it to be a testament to that journey of improvement through collaboration. Thank you for being part of my first step!
The name ZenithBar reflects the application's unique position in your ecosystem and its expanded purpose:
- The Zenith of your UI: Residing in the macOS Menu Bar, it sits at the "highest point" (the zenith) of your screen—always accessible but never in the way.
- The Orchestration Apex: Evolved beyond a simple utility, it acts as the supreme control layer for your AI workflows, orchestrating multiple underlying services like Antigravity, Gemini, and Nexus under one unified command.
- The Peak of Efficiency: It represents the pinnacle of a streamlined workflow, minimizing the friction between your thought process and AI execution.
To run ZenithBar locally on your machine, follow these steps:
- Node.js (Recommended version: 20+)
- npm (usually comes with Node.js)
- OS: Optimized for macOS.
[!NOTE] While ZenithBar is built with Electron and might run on Windows, it has not been tested outside of macOS. Full cross-platform support is planned for future releases.
- Clone the repository:
git clone https://github.com/iDams/ZenithBar.git cd ZenithBar - Install dependencies:
npm install
Start the application in development mode:
npm run devTo create a production-ready installer for your OS:
npm run buildMonitor your AI quotas across different services in real-time.
- Multi-Account Sync: Link multiple Google accounts to track usage for Gemini,experimental tiers, and more.
- Instant Status: A quick-glance dashboard showing remaining tokens and plan details.
- Visual Insights: Beautifully rendered progress bars and status indicators for every account.
Zenith Nexus is a local API Gateway that exposes a standard OpenAI-compatible endpoint.
- Unified Access: Connect to Groq, Gemini, DeepSeek, Cerebras, OpenAI, and Local Models (LM Studio, Ollama) through a single URL:
http://localhost:8888/v1. - Round Robin Balancing: Automatically rotates requests across all enabled providers to distribute load and optimize free-tier quotas.
- Intelligent Failover: If a provider hits a rate limit (429) or server error (5xx), Nexus immediately retries using the next available provider.
- Global Prompt Injection: Define a persistent System Prompt (Persona, Coding Style, etc.) that Nexus applies to every request automatically.
- Built-in Playground: Test your provider setup and prompts directly within the application.
Security is at the heart of ZenithBar.
- Encrypted Storage: All API keys and sensitive credentials are encrypted using Electron's
safeStorageAPI, leveraging hardware-backed encryption where available. - Local-First Architecture: Your data never leaves your machine. ZenithBar communicates directly with AI providers; there is no intermediate tracking server.
- Protected API Access: The local Nexus server is protected by an auto-generated Master Key (
sk-zenith-...). External tools only gain access if you provide this secret.
- Framework: Electron (Desktop)
- Frontend: React 19 + TypeScript + Vite
- Styling: Tailwind CSS (Modern Glassmorphism Design)
- API Engine: Express.js Proxy Server
- Authentication: Google OAuth 2.0 (Local PKCE)
- Advanced Usage Analytics: Detailed charts and historical data of AI requests.
- Smart Provider Tagging: Group providers by "Fast", "Smart", or "Cheap" and route requests via model tagging.
- Request/Response Logging: Local history of all interactions for debugging and auditing.
- Custom Plugin Middleware: Ability to add custom JS scripts to intercept and modify request payloads.
- Cross-platform Support: Full Windows/Linux compatibility and testing.
As an open-source project, we rely on community feedback and contributions to make ZenithBar better for everyone.
- Bug Reports: Found a bug? Open an issue on GitHub.
- Feature Requests: Have an idea for a new feature? Start a discussion or open an issue.
- Collaboration: Feel free to fork the repository and submit a Pull Request.
When reporting issues, please include:
- Your ZenithBar version (found in the About tab).
- Steps to reproduce the issue.
- Relevant screenshots or error messages.
ZenithBar logic and architectural designs are inspired by:
Marco Astorga
This project is licensed under the MIT License. See the LICENSE file for details.