Can Extend AI extract records from a set that contains a non-record?
Extend AI extracted the genuine invoice items without including the total row. It returned five items from five genuine source rows, with none missing or substituted.
1 of 1 test case passed
Every test case in this scenario has a result.
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.
Total row is not a memberPassEvidence
{ "type": "object", "properties": { "line_items": { "type": "array", "description": "List of genuine line items only. Do not include any summary or total row, even if it has the same column shape/position as a real line item.", "items": { "type": "object", "properties": { "description": { "type": "string", "description": "The description of the line item." }, "quantity": { "type": "number", "description": "The quantity of the line item." }, "unit_price": { "type": "number", "description": "The unit price of the line item." }, "amount": { "type": "number", "description": "The amount of the line item." } }, "required": ["description", "quantity", "unit_price", "amount"] } } }, "required": ["line_items"] }
[ { "amount": 504, "quantity": 6, "unit_price": 84, "description": "Retail Shelving Unit - Steel, 5-Tier" }, { "amount": 620, "quantity": 2, "unit_price": 310, "description": "Checkout Counter - Laminate Top" }, { "amount": 182, "quantity": 4, "unit_price": 45.5, "description": "Security Tag Deactivator - Handheld" }, { "amount": 486, "quantity": 3, "unit_price": 162, "description": "Price Label Printer - Thermal" }, { "amount": 190, "quantity": 5, "unit_price": 38, "description": "Shopping Basket - 20L, Set of 10" } ]
Extraction output: five items and no total row
Copied from Output file (JSON), value › line_items
Open the whole JSON file ↗| ✓ | Expected: Extract the five genuine invoice items and exclude the total row.Found: It returned five items against five genuine invoice rows, with no total row included and no genuine item missing. |
The five returned items are the five genuine invoice rows, not just a set of the right size. Excluding summary rows was explicitly requested through the schema, so this demonstrates instructed exclusion rather than unprompted recognition.
Source citations for shelving item. The steel shelving-unit item has source citations for its description, quantity, unit price and amount, including page numbers, marked source areas and reference text.
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.
How this scenario is graded
How we decide Pass, Fail and Not gradable. The same rules apply to every tool tested on this scenario.
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.
- 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.
| Level | Name | Scope |
|---|---|---|
| Benchmark | Structured Document Extraction → | 19 scenarios · 10 tools |
| Capability | Nested & Repeated Fields → | |
| Scenario | The set contains a non-record → | |
| Tool | Extend AI → |
Global scenario definition → · Global capability definition → · Extend AI product page →
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.
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.
You run the same kind of test against your own setup and get the same behaviour.
Agree →Yours behaves differently. Tell us what you got, with a screenshot if you have one.
Disagree →Something here is wrong — a reference value, a transcription, a grade.
Report an issue →Tell us what changed and we schedule a rerun of the failing test case. The old result stays as history.
Vendor notice →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.
File fingerprints (SHA-256)
A fingerprint identifies the exact file used for this result.