Track Your AI Spending Without the Dashboard Bloat

Let's be honest: managing AI API costs is becoming its own side hustle. Between ChatGPT, Anthropic, Gemini, and a dozen other services, it's easy to lose track of where your money is going each month. Most existing solutions are either locked inside a vendor's ecosystem or part of a massive, complex monitoring platform. What if you just wanted a simple, private overview?

That's where OpenUsage comes in. It's a self-hosted, minimalist dashboard built to give you a clear, centralized view of your spending across multiple AI providers. No fluff, no unnecessary features—just the numbers you need.

What It Does

OpenUsage is a straightforward web dashboard that connects to the billing APIs of various AI model providers. Once you add your API keys (it stores them locally, in your own environment), it fetches your usage and cost data, presenting it in a clean, unified interface. You can see your current month's spending, cost per provider, and a simple breakdown of your usage patterns.

Think of it as a personal finance app, but specifically for your AI API subscriptions. It answers the basic question: "How much did I spend this month, and on what?"

Why It's Cool

The appeal here is in the constraints. The project embraces minimalism by design. It doesn't try to do real-time alerting, complex forecasting, or team management. It does one job: auditing your spending. This makes it lightweight, easy to understand, and simple to host yourself.

Because it's self-hosted, your API keys and spending data never leave a server you control. For developers and indie hackers who value privacy and are cautious about linking yet another third-party service to their critical accounts, this is a significant advantage. It's built with Go and Preact, resulting in a single, efficient binary that's a breeze to deploy.

It's also the kind of tool you can easily extend. The codebase is clean and focused, making it a great starting point if you wanted to add support for a niche API provider or tweak the dashboard to show a specific metric you care about.

How to Try It

Getting started is pretty standard for a Go app. You can clone the repo and run it locally in a few minutes.

git clone https://github.com/robinebers/openusage
cd openusage
cp .env.example .env
# Edit your .env file with your API keys
go run main.go

The repository README has detailed instructions for configuration and deployment options, including running it in a Docker container. There's no live demo because the whole point is to connect it to your data, but the setup process is quick if you're comfortable with basic environment variables and running a web server.

Final Thoughts

OpenUsage fills a specific, growing need without over-engineering the solution. As AI APIs become more like utilities, having a simple way to audit their cost is just good sense. This is a perfect weekend project to deploy for yourself, and it’s a