Every developer has run into the "subscription creep" problem. You subscribe to a couple of LLM tools, an AI design generator, and a coding copilot, and suddenly you are paying $100+/month without even realizing it.
To help devs and creators audit their AI budgets, I built a lightweight price comparison database. But instead of spinning up a heavy database that would cost me money to run, I decided to go static-first.
Why Astro?
I chose Astro for https://aipriceguide.co because the site is highly content-driven. The data doesn't change by the millisecond, but it does need to load instantly.
By utilizing local markdown and JSON files for the 43+ AI tools, I kept the entire site serverless and incredibly fast. The data is parsed at build time, yielding perfect Lighthouse scores. Tailwind CSS handled the utility styling, keeping the bundle size near zero.
The Data Schema
To make comparisons reliable, I structured the tools with metadata:
- Base cost (monthly/annual)
- Token/credits model details
- Free tier limitations
- Verified free alternatives
This setup makes it incredibly easy to pull down data, run a quick build, and push updates to the live site. Check it out, and let me know if you have any questions about the build!
