Can LlamaParse repair a record by adding one member and removing one?
LlamaParse did not let the user add or remove a member of an extracted set. The saved result stayed at three line items: Site fencing remained, and Concrete mixer was absent after the edit attempts.
1 of 1 test case failed
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.
Add one member and remove oneFailEvidence
After extraction, in the Extract Results view, we tried to remove the Site fencing row and add a Concrete mixer row. We looked for row controls, double-clicked a cell, typed, pressed Delete and Backspace, and right-clicked the row.
{ "type": "object", "properties": { "line_items": { "type": "array", "description": "Every line item in the invoice table", "items": { "type": "object", "properties": { "description": { "type": "string", "description": "The description of the line item" }, "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"] }
Copied from Output file (JSON), lines 1–30
{ "file_input": "9b92246a-9a46-4465-aac7-181b782ac046", "id": "ext-j47fu0drvr2g7neyt2x5gd8yvgw6", "project_id": "205fc038-050f-4b9d-9896-b404e84e1434", "configuration_id": null, "configuration": null, "status": "COMPLETED", "error_message": null, "extract_result": { "line_items": [ { "description": "Telehandler, 4-tonne, weekly hire", "quantity": "1", "unit_price": "£850.00", "amount": "£850.00" }, { "description": "Generator, 20kVA, weekly hire", "quantity": "2", "unit_price": "£210.00", "amount": "£420.00" }, { "description": "Site fencing, per 10m panel, weekly", "quantity": "8", "unit_price": "£24.00", "amount": "£192.00" } ] },
Copied from Output file (JSON), lines 316–319
"created_at": "2026-10-28T10:32:49.939180Z", "updated_at": "2026-10-28T10:32:58.838635Z", "document_input_value": "9b92246a-9a46-4465-aac7-181b782ac046" }
Stored output: unchanged after edit attempts; last updated 8.9 seconds after extraction was created
Open the JSON file ↗| ✗ | Expected: The added member is saved, the removed member is absent, and every other member stays unchanged.Found: The saved set retained all three original line items: Site fencing remained, Concrete mixer was absent, and the other entries were unchanged. |
Adding and removing members was attempted on a successfully extracted set. The Free plan’s Extract review screen offered no way to do either, and the saved set stayed unchanged. This is an editing capability failure, not an extraction failure.
Cell editing unavailable. Double-clicking the Site fencing description selected its text; it did not open a cell editor.
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 | Review and Correction → | |
| Scenario | Repair a record → | |
| Tool | LlamaParse → |
Global scenario definition → · Global capability definition → · LlamaParse 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.