Evidence · first-party tested

Recovers dense OCR text from a scanned research page, including the report header, abstract, keywords, and opening paragraphs in one extraction.

✓ Worked🧾 artifact-verifiedTested Jun 20, 2026Adobe PDF Extract API →
Output — unretouched
OCR text from the opening page of a USDA forest service report.
OCR text from the opening page of a USDA forest service report.
The observation

Recovers dense OCR text from a scanned research page, including the report header, abstract, keywords, and opening paragraphs in one extraction.

Criterion: Text & OCR Completeness · Scenario: Scanned Research Paper (group: scanned-research-paper)
Query this
get_evidence({
  tool: "adobe-pdf-extract-api",
  scenario: "scanned-research-paper"
})
MCP · mcp.aidemos.com/api/mcp
Free with attribution.
Same input, other tools
Extend AI ⚠ StruggledLanding AI ⚠ Struggled
Real outputs, no retouching · every cell queryable via API & MCP · aidemos.com