Available Output Formats in Mathpix Python SDK

The Mathpix Python SDK allows you to convert input from images, PDFs, or Mathpix Markdown (MMD) into a wide range of output formats and download them locally.

Supported Output Formats

Format	Description	Supported endpoints
pdf	Vector PDF of the rendered content	`/v3/pdf`, `/v3/converter`
latex_pdf	PDF rendered from LaTeX instead of MMD	`/v3/pdf`, `/v3/converter`
docx	Microsoft Word file with styled math & text	`/v3/pdf`, `/v3/converter`
md	Plain UTF-8 Markdown	`/v3/pdf`, `/v3/converter`
mmd	Mathpix Markdown (source format with LaTeX)	`/v3/text`, `/v3/pdf`
html	Self-contained HTML document	`/v3/pdf`, `/v3/converter`
tex_zip	Zipped folder with LaTeX sources & images	`/v3/pdf`, `/v3/converter`
json	Line-by-line OCR output as structured JSON	`/v3/text`, `/v3/pdf`
mmd_json	Line-by-line JSON (content in MMD)	`/v3/pdf`

The to_* and to_*_file() helpers in the SDK automatically pick the correct endpoint based on your request.

Which inputs can you send to each endpoint?

Endpoint	Purpose	Supported Input Types
`/v3/text`	OCR for images and small documents	JPEG (`.jpeg`, `.jpg`, `.jpe`), PNG (`.png`), BMP (`.bmp`, `.dib`), JPEG 2000 (`.jp2`), WebP (`.webp`), Portable image formats (`.pbm`, `.pgm`, `.ppm`, `.pxm`, `.pnm`), PFM (`.pfm`), Sun Raster (`.sr`, `.ras`), TIFF (`.tiff`, `.tif`), OpenEXR (`.exr`), Radiance HDR (`.hdr`, `.pic`), GDAL-compatible raster/vector formats
`/v3/pdf`	OCR and conversion for documents	PDF, EPUB, DOCX, PPTX, AZW/AZW3/KFX (Kindle), MOBI, DJVU, DOC, WPD (WordPerfect), ODT (OpenDocument Text)
`/v3/converter`	Convert Mathpix Markdown to other formats	MMD string

Code Example

from mpxpy.mathpix_client import MathpixClient

client = MathpixClient(
    app_id="your-app-id",
    app_key="your-app-key"
)

# Convert MMD to PDF and Markdown
conversion = client.conversion_new(
    mmd="E = mc^2",
    convert_to_pdf=True,
    convert_to_md=True
)

# Wait until the conversion finishes
conversion.wait_until_complete(timeout=30)

# Save results to local files
pdf_path = conversion.to_pdf_file(path='output/equation.pdf')
md_path = conversion.to_md_file(path='output/equation.md')

# Optional: access content in memory
pdf_bytes = conversion.to_pdf_bytes()
md_text = conversion.to_md_text()

After running this script, the following files will be saved locally:

output/equation.pdf – the formatted PDF output
output/equation.md – the Markdown version of the input

Output Options Summary

Use to_*_file(path) methods to save results directly to disk.
Use to_*_bytes() or to_*_text() for in-memory access (useful for web services or databases).
You can request multiple formats in the same API call by setting multiple convert_to_* flags.
Not all formats are supported for all input types — see the full conversion spec.