All tools / Convert
PDF → JSON
Extract structured data from a PDF.
Drop a PDF here
or choose a file
Up to 50 MB
mx900-datasheet-v8.pdf
48 pages · 4.1 MB
Processing
{{ s.n }}
{{ s.label }}
{{ s.status }}
Done
17 fields extracted · 3 tables · 2 low-confidence values
{
"model": "MX900",
"family": "MX Series",
"max_input_voltage_v": 42,
"max_module_count": 16,
"operating_temp_c": [-20, 60],
"compatible_firmware": ["7.4", "8.0", "8.1"],
"_low_confidence": ["operating_temp_c"]
}
Process documents in production
POST /v1/tools/pdf-to-json
Questions
How does a custom schema work?
Describe the fields you want as a JSON shape. The extractor fills in values it finds and flags anything it could not locate confidently.
What does low confidence mean?
The value was extracted but the source was ambiguous, for example a range instead of a single number. It is still included, just flagged.
Can I see where a value came from?
Yes, the API response includes a source page for every field. This preview shows values only.