# Stowly

> Free AI hub: live models, pricing & data. First tool live — AI prompt cost calculator: paste a prompt, pick a model, see the cost before you pay. No signup.

## Tools

- [AI Token Cost Calculator](https://stowly.dev/ai-token-cost-calculator): Paste a prompt, pick a model, and see the estimated monthly cost with cache tiers, request volume, model comparison, and a cited live rate sheet.
- [Pricing](https://stowly.dev/pricing.md): Stowly is free — $0 forever, no signup, no API key. What's included, limits, and terms.
- [Auth](https://stowly.dev/auth.md): No login, no API keys, no registration. How agents use everything without credentials.
- [Waitlist](https://stowly.dev/?join=1): Join the waitlist for price history, the CLI plugin, MCP access, and new providers.

## Docs

- [About](https://stowly.dev/about.md): Why Stowly exists, what it believes, and current stats on live providers, tracked models, and build version.
- [Methodology](https://stowly.dev/methodology.md): How provider pricing becomes an estimate — cited sources, 5-minute polling, per-million-token math, cache tiers, and verification rules.
- [Sources](https://stowly.dev/sources.md): Every provider pricing page Stowly cites, with last-checked dates and per-model input/output rates per 1M tokens.

## Learn

- [Learn](https://stowly.dev/learn.md): Short explainers on the ideas behind AI costs, each ending at the calculator.
- [What is a token?](https://stowly.dev/learn/what-is-a-token.md): Plain-English explainer on subword chunks, per-million billing, and input vs output cost, with a live token visualizer.
- [What is prompt caching?](https://stowly.dev/learn/what-is-prompt-caching.md): How prompt caching works — cache reads up to 90% cheaper, write premiums, short TTLs, and when a warm prompt beats a cold one.

## Resources

- [Home](https://stowly.dev/): Live rate leaderboard across providers, the free calculator entry point, waitlist signup, and FAQ.
- [Contact](https://stowly.dev/contact.md): Contact form with topic selection (question, idea, bug, other) and a Telegram chat link.

## Legal

- [Privacy](https://stowly.dev/privacy.md): Prompts are estimated in the browser and never uploaded; no accounts; waitlist email is the only data ever collected.

## Live data

- [Pricing JSON](https://stowly.dev/api/pricing.json): Machine-readable live rates and providers with last-checked dates, refreshed every 5 minutes. Always current, unlike any snapshot.

## Key Facts

- Free AI hub for live models, pricing & data; first tool is the prompt cost calculator.
- Free, no signup, no seat pricing.
- Estimates run in the browser; prompts are never sent to Stowly servers.
- Rates are cited to provider pricing pages and re-checked every few minutes.
- Pricing math uses per-1M-token input, output, 5-minute cache write, 1-hour cache write, and cached input rates.
- Current build is v0.1, the first public launch.
- Industry: AI developer tooling, LLM cost estimation. 

## Contact

- Website: https://stowly.dev
- Telegram: https://t.me/stowlydev
- Contact page: https://stowly.dev/contact