LYRENTH
HomePricingIntegrationsDocsBlogBenchmarksIndex statsAboutContact
Benchmarks

What famous docs pages cost an agent, raw vs indexed.

Every row below is a real page read through the Lyrenth index, with the token economics reported by the API itself: what the raw HTML would cost a model to ingest, what the clean AIDocument costs instead, and the difference. Measured 2026-07-05. Original data, reproducible with one API call.

PageRaw HTML tokensAIDocument tokensSmallerSaved
Stripe API reference
Payments
307,9022,000154x99.4%
Vercel Functions docs
Serverless
237,3902,149110x99.1%
Cloudflare Workers docs
Serverless
94,9631,37769x98.5%
GitHub REST API quickstart
Developer platform
88,6144,87618x94.5%
Kubernetes Pods concepts
Infrastructure
131,9777,47518x94.3%
React useState reference
Frontend
111,0428,15614x92.7%
MDN: Overview of HTTP
Web fundamentals
56,6925,08511x91.0%
Python asyncio docs
Languages
6,1751,2514.9x79.7%
Node.js fs module
Languages
259,563102,3722.5x60.6%
PostgreSQL SELECT reference
Databases
29,50917,5001.7x40.7%

The spread is honest: dense reference pages like PostgreSQL's SELECT save less, documentation apps wrapped in heavy chrome save the most. Cost basis: $3.00 per 1M input tokens (Claude Sonnet / GPT-4 Turbo class), as reported in the API's economics block.

Get API access Free tier · no credit card