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

Can Datalab extract repeated records while keeping each record’s fields together?

Datalab extracted the repeated set without mixing values between items. All three invoice items were returned, with nothing dropped or duplicated.

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.

Repeated set with correct associationPassEvidence
What we sent

We ran Forge Extract once on the three-page invoice, using Balanced mode and an extraction schema for the items’ descriptions, product codes, quantities, unit prices and amounts.

Input 1Document (PDF)
Input PDF: three-page invoice with the goods table on page 3
Open the PDF ↗
Input 2File (JSON)
Extraction schema: the goods table’s five requested properties
JSON
{
  "type": "object",
  "properties": {
    "line_items": {
      "type": "array",
      "description": "Every line item in the goods table on page 3",
      "items": {
        "type": "object",
        "properties": {
          "description": {
            "type": "string",
            "description": "The description of the line item"
          },
          "item_code": {
            "type": "string",
            "description": "The item or product code printed for the line"
          },
          "quantity": {
            "type": "string",
            "description": "The quantity of the line item"
          },
          "unit_price": {
            "type": "string",
            "description": "The unit price of the line item"
          },
          "amount": {
            "type": "string",
            "description": "The amount of the line item"
          }
        }
      }
    }
  },
  "required": ["line_items"]
}
Open the JSON file ↗
What the tool returned
JSONlines 1–265 of the file · copied lines 3–263
{
  "line_items": [
    {
      "description": "Dosing pump, 240V",
      "description_citations": [
        {
          "block": "/page/2/Table/3",
          "page": 2,
          "text": "Dosing pump, 240V",
          "offset": 39,
          "length": 17,
          "score": 1,
          "bbox": [151, 347, 346, 376]
        }
      ],
      "item_code": "DP-2210",
      "item_code_citations": [
        {
          "block": "/page/2/Table/3",
          "page": 2,
          "text": "DP-2210",
          "offset": 57,
          "length": 7,
          "score": 1,
          "bbox": [742, 347, 826, 374]
        }
      ],
      "quantity": "2",
      "quantity_citations": [
        {
          "block": "/page/2/Table/3",
          "page": 2,
          "text": "2",
          "offset": 65,
          "length": 1,
          "score": 1,
          "bbox": [1018, 347, 1032, 374]
        }
      ],
      "unit_price": "£340.00",
      "unit_price_citations": [
        {
          "block": "/page/2/Table/3",
          "page": 2,
          "text": "£340.00",
          "offset": 67,
          "length": 7,
          "score": 1,
          "bbox": [1174, 347, 1252, 374]
        }
      ],
      "amount": "£680.00",
      "amount_citations": [
        {
          "block": "/page/2/Table/3",
          "page": 2,
          "text": "£680.00",
          "offset": 75,
          "length": 7,
          "score": 1,
          "bbox": [1367, 349, 1442, 376]
        }
      ]
    },
    {
      "description": "Chemical feed tubing, 10m coil",
      "description_citations": [
        {
          "block": "/page/2/Table/3",
          "page": 2,
          "text": "Chemical feed tubing, 10m coil",
          "offset": 83,
          "length": 30,
          "score": 1,
          "bbox": [153, 412, 454, 439]
        }
      ],
      "item_code": "CT-1040",
      "item_code_citations": [
        {
          "block": "/page/2/Table/3",
          "page": 2,
          "text": "CT-1040",
          "offset": 114,
          "length": 7,
          "score": 1,
          "bbox": [742, 412, 823, 439]
        }
      ],
      "quantity": "3",
      "quantity_citations": [
        {
          "block": "/page/2/Table/3",
          "page": 2,
          "text": "3",
          "offset": 122,
          "length": 1,
          "score": 1,
          "bbox": [1018, 412, 1032, 439]
        }
      ],
      "unit_price": "£48.00",
      "unit_price_citations": [
        {
          "block": "/page/2/Table/3",
          "page": 2,
          "text": "£48.00",
          "offset": 124,
          "length": 6,
          "score": 1,
          "bbox": [1185, 412, 1251, 439]
        }
      ],
      "amount": "£144.00",
      "amount_citations": [
        {
          "block": "/page/2/Table/3",
          "page": 2,
          "text": "£144.00",
          "offset": 131,
          "length": 7,
          "score": 1,
          "bbox": [1367, 412, 1442, 439]
        }
      ]
    },
    {
      "description": "Inline flow sensor",
      "description_citations": [
        {
          "block": "/page/2/Table/3",
          "page": 2,
          "text": "Inline flow sensor",
          "offset": 139,
          "length": 18,
          "score": 1,
          "bbox": [151, 474, 322, 504]
        }
      ],
      "item_code": "FS-3302",
      "item_code_citations": [
        {
          "block": "/page/2/Table/3",
          "page": 2,
          "text": "FS-3302",
          "offset": 158,
          "length": 7,
          "score": 1,
          "bbox": [742, 474, 821, 501]
        }
      ],
      "quantity": "1",
      "quantity_citations": [
        {
          "block": "/page/2/Table/3",
          "page": 2,
          "text": "1",
          "offset": 166,
          "length": 1,
          "score": 1,
          "bbox": [1018, 474, 1032, 504]
        }
      ],
      "unit_price": "£215.00",
      "unit_price_citations": [
        {
          "block": "/page/2/Table/3",
          "page": 2,
          "text": "£215.00",
          "offset": 168,
          "length": 7,
          "score": 1,
          "bbox": [1176, 477, 1252, 501]
        }
      ],
      "amount": "£215.00",
      "amount_citations": [
        {
          "block": "/page/2/Table/3",
          "page": 2,
          "text": "£215.00",
          "offset": 168,
          "length": 7,
          "score": 1,
          "bbox": [1176, 477, 1252, 501]
        }
      ]
    }
  ]
}

Returned JSON: three extracted items with values and source citations

Copied from Output file (JSON), lines 3–263

Open the JSON file ↗
Expected vs. Found
✓Expected: Return all three items, with the right values attached to each item and no omissions or duplicates.Found: The invoice has three goods rows; Datalab returned three items, preserving each row’s values together without dropping or duplicating items.
Supporting proof
Proof 1Screenshot
Results screen: items 1 and 2 beside the invoice
Open original ↗
Proof 2Screenshot
Results screen: item 2’s remaining values and item 3
Open original ↗
Proof 3Screenshot
Configuration screen: Balanced mode and the required array with five text properties
Open original ↗
Why this result

Datalab’s extraction feature met the repeated-set requirement by keeping each item’s values together, with none lost or duplicated. This result covers Balanced mode.

Also observed in this test case

Flow sensor amount citation. For the inline flow sensor, the amount’s citation points to the unit price cell, not the amount column. Both cells contain £215.00.

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
The task required extracting the invoice’s page 3 goods table—dosing pump, chemical feed tubing and inline flow sensor—with each item’s description, product code, quantity, unit price and amount.
Extraction 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
CapabilityNested & Repeated Fields →
ScenarioA repeated set of records →
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
20 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-repeated-set-of-records · 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 · 5 KB
Proof 1 · Screenshot ↗image/png · 97 KB
Proof 2 · Screenshot ↗image/png · 92 KB
Proof 3 · Screenshot ↗image/png · 72 KB
File fingerprints (SHA-256)

A fingerprint identifies the exact file used for this result.

Input 1 · Document (PDF)89fb14d7c0897bbfef47ffa3aa3930ee7eccf62c8a9430710d1906598c9bdfec
Input 2 · File (JSON)7c444fd82f3362e9df41c7ef13044a3374bd44c72b6c68e593ec37cc3e44c7db
Output file (JSON)6843bd75c8c23714e3f3573361fc09484057aa3488043e84d30437d48b5a2e31
Proof 1 · Screenshot8ca043ae82a21dfa63b5b45c50567afb06ff5b131f5d85f9532bc0188ddf9c44
Proof 2 · Screenshot5ef675f6d43e1e0d110b1d8d8ccf1dc931aeff46466eeb7ee378cbe64de3006c
Proof 3 · Screenshotc41e162d9a1bd1cee6f751e324b5f566f0255a3ef8bd1e248c99fab6d31fe33a