AI Demos Research — the structured-intelligence platform. Every verdict on these pages opens to the execution behind it.
Graded 19 November 2026

Does Datalab read the scanned pages of a document that mixes digital and scanned pages?

Datalab returned all five requested values from the three-page invoice, including the vehicle registration and the recipient's name that appear only on its scanned page. Each value matched the document and was cited to the page it sits on, and the total due kept its pound sign.

1 of 1 test case passed

Every test case in this scenario has a result.

Pass rate100%1 of 1 gradable
Coverage1 of 1test case with a result
1PassDid everything it was expected to do.
0FailNo test case failed.
0Not gradableEvery result here could be graded.
0UntestedEvery test case in this scenario has a result.
The pass rate is a summary. The evidence is the test case below: what we sent, what we checked, what the tool returned, and the proof.

The test case

Each test case is judged on its own: Pass, Fail, Not gradable, or Untested. The scenario result above is based on this test case.

Values on both digital and scanned pagesPassEvidence
What we sent

In the Datalab Forge playground, we ran the Extract step once in Balanced mode on the three-page invoice, with a five-field schema.

Input 1Document (PDF)
Input PDF: three-page invoice, with an image-only scanned delivery note as page 2
Open the PDF ↗
Input 2File (JSON)
Input schema: the five requested fields, all text, none required
JSON
{
  "type": "object",
  "properties": {
    "invoice_reference": { "type": "string", "description": "The invoice reference number printed on the invoice" },
    "order_reference": { "type": "string", "description": "The order reference the invoice relates to" },
    "vehicle_registration": { "type": "string", "description": "The registration plate of the delivery vehicle" },
    "received_by": { "type": "string", "description": "The name of the person who received the delivery" },
    "total_due": { "type": "string", "description": "The total amount due on the invoice" }
  }
}
Open the JSON file ↗
What the tool returned
JSON
{
  "invoice_reference": "HIS-4402",
  "order_reference": "FWT-PO-2291",
  "vehicle_registration": "LR19 KVM",
  "received_by": "J. Whitlock",
  "total_due": "£1,246.80"
}

Returned JSON: the five values, each with a citation to its page

Copied from Output file (JSON)

Open the JSON file ↗
Expected vs. Found
✓Expected: Every requested value returned, from the digital pages and the scanned page alike, with none missing, invented or degraded.Found: All five returned: invoice and order references from page 1, registration LR19 KVM and recipient name from scanned page 2, £1,246.80 from page 3, each cited to its page.
Supporting proof
Proof 1Screenshot
Screenshot: the results screen with all five fields and their values
Open original ↗
Proof 2Screenshot
Screenshot: the Extract step set to Balanced mode with the five-field schema
Open original ↗
Why this result

The test required values from both the digital pages and the image-only scanned page, with nothing missed or invented. The two values found only on the scanned page came back matching the delivery note, and the digital values came back intact, each cited to its own page.

Tested by Rugved Nichite · evidence dated 19 November 2026

Configuration and setup

How this tool was set up for the run and what the test needed in place. Each row is a fact from the run's records; a fact the records do not hold is left out, not guessed.

Software that produced the output
Datalab
Build or version
Not exposed by the hosted product
Surface
Web app
Set up before the run
A three-page invoice: digital page 1 with the invoice and order references, an image-only scanned delivery note as page 2 holding the vehicle registration and recipient, and digital page 3 with the total due. All five values were requested.
Extract mode
Balanced
Tested
19 November 2026 · Rugved Nichite

How this scenario is graded

How we decide Pass, Fail and Not gradable. The same rules apply to every tool tested on this scenario.

How results are decided

Each test case gets one result per tool: Pass, Fail or Not gradable. A test case we haven't run yet shows Untested. There are no partial results.

The rules
  • Pass: the tool did everything the test expected, and nothing it said contradicts the correct answer.
  • Fail: at least one expected behaviour clearly didn't happen; the test case's result says which one and quotes the tool.
  • Not gradable: our evidence couldn't settle the outcome (for example a record we needed is missing). It is never counted as a fail, and the test case's result says what's missing.

Where this sits in the benchmark

This page is one cell of a larger study: one tool, one scenario. Only this benchmark's frame appears here.

LevelNameScope
BenchmarkStructured Document Extraction →19 scenarios · 10 tools
CapabilityOCR →
ScenarioA document mixing digital and scanned pages →
ToolDatalab →

History of this result

What has happened to this result since it was first published. Runs and grades are never overwritten: a retest or a re-grade publishes a new result and keeps the earlier one readable.

from the publication record
19 November 2026First publishedStructured Document Extraction v1

Act on this result

Nothing filed here edits the run or the grade. A challenge opens a review, and a review can produce a new run or a re-grade — which becomes the current result and leaves this one in the history.

This matches what I see

You run the same kind of test against your own setup and get the same behaviour.

Agree →
This does not match

Yours behaves differently. Tell us what you got, with a screenshot if you have one.

Disagree →
Point out an issue

Something here is wrong — a reference value, a transcription, a grade.

Report an issue →
Request a retest

On a newer build, a larger dataset, or your own setup.

Request a retest →
We have fixed this

Tell us what changed and we schedule a rerun of the failing test case. The old result stays as history.

Vendor notice →
Filed against this evidenceNothing yet. Challenges, counter-evidence and fix notices appear here with their outcome, and stay on the page after they are resolved.
Cite this result
aidemos.com/benchmarks/structured-document-extraction/results/datalab/a-document-mixing-digital-and-scanned-pages · 1 pass · 0 fail · coverage 1/1 · graded 2026-11-19

The same record is available as structured data through the AI Demos MCP server, with the counts, the coverage and every per-test-case reason carried as fields.

Verify the proof files

These files support this result. Open a file to inspect the original evidence.

Input 1 · Document (PDF) ↗application/pdf · 176 KB
Input 2 · File (JSON) ↗application/json · 1 KB
Output file (JSON) ↗application/json · 3 KB
Proof 1 · Screenshot ↗image/png · 59 KB
Proof 2 · Screenshot ↗image/png · 93 KB
File fingerprints (SHA-256)

A fingerprint identifies the exact file used for this result.

Input 1 · Document (PDF)89fb14d7c0897bbfef47ffa3aa3930ee7eccf62c8a9430710d1906598c9bdfec
Input 2 · File (JSON)30ee573d0c27218f97f6f31990de28f0eda02281ee3da3ffe5bc8ebe4c9a5dbc
Output file (JSON)5b7d75f73e1b182662a7cc8c72ede30bf1bc873839da9ac6af5b7ed6d7234e88
Proof 1 · Screenshotbf25b89748a444fa17ef8d86aef343bf99b5d516307435ef13796f4d8151eefc
Proof 2 · Screenshot466c3be34e709713b9e86a2747a4d458ffb2f06cc3f9056a4828fb4754448293