Tools that show their work.
We build small, fast, single-purpose AI tools — and explain exactly how each reading is calculated, so you can judge how much to trust it.
Why this site exists
Most "AI detector" and "AI cost calculator" tools online are either paywalled after the first use, vague about their methodology, or both. We started this site to build the opposite: free, fast, single-purpose tools that are upfront about what they measure, what they can't prove, and how the number on screen was produced.
What we build
Every tool on this site runs a specific, narrow job well: detecting likely AI-written text, humanizing stiff phrasing, counting tokens across model families, estimating API spend, and organizing prompts you actually reuse. We'd rather ship eleven tools that each do one thing clearly than one tool that tries to do everything and explains none of it.
How we think about accuracy
AI text detection in particular is a probabilistic exercise, not a lie detector. Our detection tools report a likelihood, describe the signals behind that likelihood — sentence-length variation, word predictability, phrasing patterns — and are explicit that no detector, including ours, should be the sole basis for an accusation of academic or professional misconduct. Every detection tool page includes a plain-language guide to what the score does and doesn't mean.
Privacy by default
Wherever a tool can run entirely in your browser, it does — text you paste into the token counter, cost calculator, or context visualizer never leaves your device. Tools that need a server call for higher-quality results (like the AI Humanizer's advanced mode) say so directly next to the input box. See our Privacy Policy for details.
Get in touch
Found an inaccurate reading, a broken tool, or have a tool you'd like us to build? Contact us — we read every message.