Documents you can trust.
tdoc turns PDFs, DOCX, and HTML into typed, queryable, cryptographically‑signed documents. Every number carries a declared type. Every archive hashes to the same bytes, always. Ship it in one REST call.
Not a mock. Real bytes. From our own production API.
Request ON LOAD
GET https://api.tdoc.xyz/v1/healthz
accept: application/json
user-agent: tdoc-landing/0.1
# api.tdoc.xyz is a Cloudflare Worker proxying to a
# Hugging Face Space running product/service/main.py.
# Uptime: since 2026-04-24.
Response PENDING
A format, not a file.
AXON separates what a document says from how it renders.
The content tree is typed. A pvalue
is a p-value. A measure carries units.
A currency knows its ticker.
Your agent never has to guess whether “p = 0.03” is a probability, a typo, or a page number. It’s declared. It’s signed. It round-trips to the same bytes.
<section id="results">
<heading level="2">Outcome</heading>
<paragraph>
<measure type=pvalue
value=0.003>p = 0.003</measure>
at the 10th minute, with a
<measure type=ci
value=[0.41, 0.78]>
95 % CI [0.41, 0.78]
</measure>.
</paragraph>
<table id="t1" summary="arm-by-arm">
<cell type=currency
ticker="USD">$12,400</cell>
<cell type=measure
unit="mg/dL">142.7</cell>
</table>
</section>
The ledger.
data-type: pvalue, measure, currency, ci, date, &c. Your downstream code never guesses.validate(), not bolted on.Prices, plain.
No credit card · No seat taxes · No surprises
-
Free· · · · · · · · · · · · · · · · · · · ·100 documents / month. All parsing features. AQL queries included. Forever.$0/mo
-
Pro· · · · · · · · · · · · · · · · · · · ·2,000 documents / month. Email support. Upgrade anytime — no hard overage block.$29/mo
-
Team· · · · · · · · · · · · · · · · · · · ·10,000 documents / month. Ed25519 signing & verification. 24-hour email SLA. Overage $0.02/doc.$99/mo
-
Scale· · · · · · · · · · · · · · · · · · · ·100,000 documents / month. Self‑hosted option. Slack support. Overage $0.01/doc.$499/mo
Every AI team is losing 20–40 % of document quality to broken PDF ingestion. Tables become strings. Equations become gibberish. P‑values lose their type. tdoc gives agents the same structured view a database has — so your models cite a number, rather than merely see it.