Can LlamaParse aggregate across documents?
LlamaParse did not produce the required combined total of 474.93 across three receipts in Extract on the Free plan. It filled the requested cross-document sum with individual receipt totals instead.
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.
Total across every documentFailEvidence
We uploaded three receipt PDFs—the baseline receipt, the invoice with a different total label, and the different-layout receipt—to Extract, one at a time. Each used the same schema, including a field requesting the sum of the total due across all documents in the project.
Sum of the total due across all of the documents uploaded in this project
Copied from Output file 1 (JSON), line 14
"total_due_all_documents": "$90.10"
Copied from Output file 1 (JSON), lines 107–131
"total_due_all_documents": { "citation": [ { "page": 1, "matching_text": "$90.10", "bounding_boxes": [ { "x": 512.6664632463455, "y": 304.2129379463196, "w": 31.964370346069337, "h": 10.385311059951782 } ], "page_dimensions": { "width": 594.96, "height": 841.92 } } ], "confidence": 0.9420762965080913, "extraction_confidence": 0.9420762965080913 } }, "page_metadata": null, "row_metadata": null
Copied from Output file 2 (JSON), line 14
"total_due_all_documents": "$227.33"
Copied from Output file 3 (JSON), line 14
"total_due_all_documents": "USD 157.50"
Returned JSON: the baseline receipt’s cross-document sum points to its own total
Open the JSON file ↗Returned JSON: the different-label invoice’s cross-document sum points to its own total
Open the JSON file ↗Returned JSON: the different-layout receipt’s cross-document sum points to its own total
Open the JSON file ↗| ✗ | Expected: The correct total across every extracted document: 474.93, summed from the amounts the product returned.Found: The combined-total field returned $90.10, $227.33 and USD 157.50 in separate answers, rather than the required 474.93 sum of the extracted amounts. |
Extract read each receipt correctly but offered no option to query them together. Because a combined answer was required inside the product, this is a querying failure rather than an extraction error, limited to Extract on the Free plan.
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 | Querying → | |
| Scenario | Aggregate across documents → | |
| 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.