◇ BenchmarkServerless
Vercel Functions docs: 99.1% fewer tokens.
Vercel's serverless functions guide, a heavy modern docs app. Read through the Lyrenth index, the page becomes one clean AIDocument of 2,149 tokens instead of 237,390 tokens of raw HTML: 110x smaller for every agent that reads it. Numbers straight from the API's economics block, measured 2026-07-05.
Raw HTML · 237,390 tokens
AIDocument · 2,149 tokens
| Cost to ingest raw HTML | $0.712 |
| Cost to ingest the AIDocument | $0.0064 |
| Saved, per read, per agent | $0.706 |
Basis: $3.00 per 1M input tokens (Claude Sonnet / GPT-4 Turbo class), as reported in the API's economics block. Source page: https://vercel.com/docs/functions
Reproduce it
One call, same numbers.
Every AIDocument response carries its own economics block, so this benchmark reproduces itself:
curl -X POST https://api.lyrenth.com/v1/aidocument \
-H "Authorization: Bearer $LYRENTH_KEY" \
-H "Content-Type: application/json" \
-d '{"url":"https://vercel.com/docs/functions"}'