We tested hosted PDF-to-markdown APIs on the same three hard documents: a long hybrid annual report, a table-heavy financial report, and an image-only scanned research paper. The goal was usable markdown with OCR, tables, charts, and reading order preserved well enough for downstream RAG, search, and reuse.
Tied with Extend AI on hybrid reports; best for programmatic chart extraction; scanned tables drop performance.
The scan text stayed readable on both dense prose and a blurred stamp, but the evidence is narrower than for a full 5 because there is not broad proof across many degraded pages.
The evidence-backed checks show the shape of the field; coverage explains the gaps.
Columns, left to right: Chart retention · Image retention · OCR quality on scans · Reading order & structure · Separate table/chart extraction (bonus) · Table structure preservation · Transparency
Pick the tools you care about, then compare what they returned or how they scored.
It handled the 84-page hybrid annual report very well: the reading order stayed coherent, tables were preserved, and even charts, signatures, and logos were represented, though mostly as text rather than inline visuals.
llamaparse-earnings-hybrid-pdf-input-page-3-6bbbf1e08dfb.png
It handled the annual report well: the page flow stayed intact, the financial table was rebuilt clearly, and the waterfall chart was surfaced separately, but the signature visuals were not kept as images and the blurry stamp had a small OCR slip.
Written result only
It keeps the earnings report readable from headings through tables and charts, and it even carries along signatures and stamps, but the logo is only partially faithful as an image.
Written result only
It turned the earnings report into usable markdown with the main table, waterfall chart, signatures, and blurry stamp all represented, but the top-level page heading lost its heading treatment.
Written result only
It handled the long annual report well overall: the main section flow stayed in place, the financial table was rebuilt clearly, and page assets were kept with the right pages, but some headings were flattened in places.
Written result only
It handled the main financial table and chart, but the two-column narrative and the signatures page lost structure, so the document is only partly faithful overall.
Written result only
It handled the page flow and some text well, but the table, waterfall chart, and handwritten signatures were all weakened or dropped, so the overall result is only fair.
Written result only
No output file was captured for this test.
Written result only
Open a tool to inspect every recorded check and finding.
The charts are not kept as pictures, but their values and labels are carried over faithfully, which is strong enough for a high score without being a perfect visual retention result.
Scored, but no finding or artifact was recorded. We do not present the number as proof.
LlamaParse is the page’s winner, and that matches the scorecard: it leads on hierarchy/reading order, table handling, and scanned OCR, which are the decisive strengths on this page. The caveat is that it is only partly tested (4 of 10 decisive checks), so the evidence is narrower than for some lower-ranked tools; still, by the published rank order it is the overall pick. If you need document flow plus ordinary table reconstruction, Tensorlake is the next fit, but it is weaker on hierarchical tables and image retention. Extend AI is a solid alternative for preserving flow and charts, but its complex-table and image-handling results are less dependable. Landing AI stands out for chart/table recovery, yet its weaker headings and scanned OCR make it less balanced. Upstage AI is best when the goal is extracting tables and charts into usable text, but it does not preserve page layout as well. Adobe API has no decisive checks here, so there is not enough evidence to place it meaningfully.
The tools we tested for this use case — each card opens its full tested review.
If you are looking to build a custom PDF-to-markdown conversion, OCR extraction, or document parsing system for your business or internal workflow, email us at contact@futuresmart.ai.
Found something inaccurate or missing? We try to keep our AI research accurate and useful. If you found outdated information, an issue, or have a suggestion, email us at collaborate@aidemos.com.
Comments (0)