Does Nutrient.io preserve styled headings and their nesting in Markdown?
Nutrient.io's Build API produced real Markdown headings but did not preserve their nesting. It put subsections at the same level as their parent sections and turned two main sections into lower-level headings.
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.
Styled headings — hierarchy preservedFailEvidence
We sent the complete 3-page laboratory sample-reception PDF to Nutrient.io's Build API for conversion to Markdown in one request.
Copied from Output file (Markdown), lines 34–42
# Condition on arrival Reception staff record the condition of each container on the submission form before it leaves the reception bench. Anything recorded there that falls outside the limits below goes to the duty chemist before the sample is booked in. # Temperature
Copied from Output file (Markdown), lines 58–68
# Storage # Refrigerated storage Samples for chemical analysis are held at 4 degrees in the reception cold room and are racked by the day of receipt, not by client. Racks are cleared every Friday afternoon. # Frozen storage
Copied from Output file (Markdown), line 86
## Records
Copied from Captured file (JSON), response
{ "status_code": 200, "headers": { "date": "Wed, 09 Sep 2026 08:51:57 GMT", "content-type": "text/markdown", "content-length": "7693", "content-disposition": "attachment; filename=\"result.md\"", "server": "Cowboy", "x-pspdfkit-remaining-credits": "47", "x-pspdfkit-request-cost": "1", "x-request-id": "GNObA4Z5kpAMVskAAKCy" }, "content_type": "text/markdown", "body_bytes": 7693, "delivery": "raw markdown body (content-type text/markdown)", "markdown_chars": 7691, "zip_members": null, "asset_names": null }
Returned Markdown: sections and subsections placed at the same heading level
Open the Markdown file ↗API exchange: successful Markdown response and reported credit usage
Open the whole JSON file ↗| ✗ | Expected: Preserve the source's three-tier heading hierarchy as real Markdown headings, not plain or bold text.Found: The source's three visual tiers were lost: 24 of 26 Markdown headings were level 1, two level 2, and none level 3. |
- request
- { "method": "POST", "url": "https://api.nutrient.io/build", "multipart_fields": ["instructions", "file"], "instructions": { "parts": [ { "file": "file" } ], "output": { "type": "markdown" } }, "input_file": "procedure_KAL-SP-06_sample_reception.pdf", "input_bytes": 43662 }
The supported Build API conversion completed, but real Markdown headings alone did not meet the requirement to preserve nesting. This conclusion concerns its default pipeline. Nutrient.io also documents a separate PDF-to-Markdown API, which merged a section and subsection in a separate test.
Body prose retained. The returned Markdown retained the body prose under every heading.
Credit usage reported. For this conversion, the Build API reported one credit billed and 47 credits remaining on the account.
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 | Converting a complex PDF into clean Markdown with a hosted API → | 12 scenarios · 13 tools |
| Capability | Heading & Section Structure → | |
| Scenario | A document with styled headings and subheadings → | |
| Tool | Nutrient.io → |
Global scenario definition → · Global capability definition → · Nutrient.io 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.