PDF to Markdown
Convert text-based PDFs into agent-readable Markdown.
The PDF to Markdown API accepts exactly one public PDF URL. It returns Markdown plus extracted image files when the PDF contains figures or embedded images.
Endpoint
POST https://agent.aetherlab.me/v1/convert/pdf_md
Request
curl -X POST https://agent.aetherlab.me/v1/convert/pdf_md \
-H "content-type: application/json" \
-H "sk: sk_live_xxxxxxxxxxxxx" \
-d '{"files":["https://example.com/report.pdf"]}'
Response
urls is an array. Most conversions return one Markdown URL. PDFs with images can return the Markdown file plus image URLs.
{
"jobId": "d245e257-a179-48c5-bf0a-92a391977384",
"feature": "pdf_md",
"urls": [
"https://agent.aetherlab.me/converted/2026-07-16/report-3bdce9696dac89bd531aedc65c45db3c.md",
"https://agent.aetherlab.me/converted/2026-07-16/report-3bdce9696dac89bd531aedc65c45db3c-IMG1.png",
"https://agent.aetherlab.me/converted/2026-07-16/report-3bdce9696dac89bd531aedc65c45db3c-IMG2.png"
]
}
Notes
- The input PDF must be reachable through a public HTTP or HTTPS URL.
- Text-based PDFs work best. Scanned image-only PDFs are not OCR targets.
- Each request consumes one credit.
