v1.0 / online

Query any PDF
at terminal
speed.

Jack into a command-line oracle for your documents. Upload a PDF, ask anything, get precise answers with exact page citations. 400-page whitepapers, parsed in one query.

Start queryingView protocol
→ No credit card→ 7-day money-back→ Cancel anytime
~/dochat/whitepaper.pdf● LIVE
Doc / 312 pages
Query session
> user
// live transcript / cycling

$ dochat query --doc

$ ask
"What's the termination clause in this contract?"
→ response
Section 14.2 grants either party the right to terminate with 30 days' written notice. Page 24.
cite: p.24
$ ask
"Compare the EBITDA margins for FY22 vs FY23."
→ response
FY22 EBITDA margin: 18.4%. FY23: 21.7%. A 330 bps expansion driven by cost efficiencies.
cite: p.41, p.42
$ ask
"List every compliance deadline in this report."
→ response
I found 7 deadlines: SOC2 audit (Mar 31), GDPR review (Apr 15), ISO renewal (May 20)…
cite: p.78, p.79, p.80
// 04 use cases

Built for anyone buried in PDFs.

01

Research papers

Extract methodology, datasets, and conclusions from dense academic PDFs in seconds.

02

Legal contracts

Surface termination clauses, liability caps, and deadlines without re-reading 80 pages.

03

Financial reports

Query 10-Ks, earnings decks, and investor memos for specific line items.

04

Technical manuals

Debug hardware specs, config flags, and API references with exact page refs.

// 03 protocol

Three commands. One answer.

01step / 1 of 3
$ upload --file doc.pdf

Upload

Drop any PDF — contracts, papers, reports. Up to 500 pages, any layout.

02step / 2 of 3
$ ask 'what are the key terms?'

Ask

Type a question in plain English. We parse it, vectorize the doc, and retrieve the best chunks.

03step / 3 of 3
$ read_response --cite pages

Receive

Get streaming answers grounded in the document, with exact page citations you can click to verify.

// 05 features

Everything you need. Nothing you don't.

Instant answers

Streaming responses powered by Claude Haiku. First token in under 500ms.

Page citations

Every claim links back to the exact page. Zero hallucinations tolerated.

Chat history

Revisit every conversation. Threaded by document, searchable by keyword.

Multiple PDFs

Upload as many documents as your plan allows. Switch context instantly.

Secure storage

Your files are encrypted at rest. Only you can access them. Deleted on demand.

API access

Business tier ships a REST API to embed Dochat in your own tools.

// 06 operator feedback

Used by people who read too much.

★★★★★ / 5.0
“I stopped reading 60-page contracts cover to cover. Dochat surfaces the clauses I need in under 10 seconds. Our team saved 40 hours in the first month.”
Marcus Avery
Partner, Hawthorne & Frye LLP
@mavery
★★★★★ / 5.0
“As a researcher I used to CTRL+F through hundreds of PDFs. Now I just ask. The page citations are accurate enough that I cite them directly in my lit reviews.”
Dr. Imani Odera
Biomedical researcher, MIT
@iodera
★★★★★ / 5.0
“Our analyst team parses 30 earnings reports per week. Dochat became a drop-in replacement for manual skimming. The API access on Business is the cherry on top.”
Kenji Park
Head of Research, Northstar Capital
@kpark
// 07 pricing

Simple pricing.

// no free tier. no lorem. just two real plans.

Professional

$15/mo

Billed monthly

For individuals parsing a handful of docs per month.

  • 10 PDFs uploaded per month
  • 500 pages per document
  • Unlimited questions
  • Page citations on every answer
  • Full chat history
  • Email support
Get Professional
// recommended
Business

$39/mo

Billed monthly

For teams with unlimited documents and API access.

  • Unlimited PDFs
  • Unlimited pages per document
  • Priority inference queue
  • REST API access
  • Team seats (coming soon)
  • Priority support + SLA
Get Business
// 08 faq

Questions.

Every answer is grounded in the specific text chunks we retrieve from your PDF. We cite the exact page so you can verify. If the answer isn't in the document, Dochat will say so explicitly instead of making things up.
PDF only for now. We extract both text-layer PDFs and (in beta) scanned documents via OCR. Images and tables are preserved in the retrieval context. We plan to add .docx, .epub, and URL ingestion next.
Your files are encrypted at rest and only accessible with your credentials. We never train on your documents. You can delete any file at any time and it is purged from storage within minutes.
Professional plan: 500 pages per PDF, 10 PDFs per month. Business plan: unlimited. For enterprise volumes (10k+ pages per doc) get in touch for a custom deployment.
Yes. Cancel from your dashboard and keep access until the end of the billing period. We also offer a 7-day money-back guarantee on your first payment, no questions asked.