Mathpix Convert vs GLM-OCR
Pick Mathpix Convert for math-heavy papers and textbooks, handwritten equations, and PDFs you need as LaTeX or DOCX. Pick GLM-OCR for high-volume PDF to Markdown at low cost, or when you want to self-host the model. The main split is STEM output formats and hosted-only access versus a cheap, MIT-licensed model that you can run yourself but that has known table and repetition bugs.
Mathpix Convertmathpix/convert | GLM-OCRzai/glm-ocr | |
|---|---|---|
| Type | OCR API | Open source |
| Price | 7 credits / page | 6 credits / page |
| Per 1,000 pages | $7.00 | $6.00 |
| Typical speed | 1.5 s / page | 1.3 s / page |
| Best for | Math and scientific papers | Low-cost general documents |
| Inputs | PDF, images | PDF, images |
| Languages | 30+ languages | 100+ languages |
| Max pages per request | 500 | 500 |
| Tables | ✓ Yes | ✓ Yes |
| Handwriting | ✓ Yes | No |
| Math | ✓ Yes | ✓ Yes |
| Bounding boxes | ✓ Yes | ✓ Yes |
| Confidence | ✓ Yes | ✓ Yes |
| Forms | No | No |
| Multilingual | No | ✓ Yes |
| On OCRRouter | Available | Available |
Prices are what you pay on OCRRouter. Speed is the provider's typical figure, not a benchmark we ran.
Which to choose
Choose Mathpix Convert if
- You need LaTeX or DOCX output from whole PDFs, which GLM-OCR does not produce (it outputs Markdown, JSON and plain text).
- Your documents are math-heavy papers or textbooks with equations, tables and chemistry, or contain handwritten equations.
- You want a hosted API only and do not need to run a model yourself.
Choose GLM-OCR if
- You process PDFs in volume and want the lowest cost: $0.03 per million tokens on the Z.ai API, or free weights you run on your own compute.
- You need to self-host under an MIT licence with vLLM, SGLang, Ollama or transformers.
- You want to extract fields from forms into a user-defined JSON schema, not just transcribe text.
Key differences
| Mathpix Convert | GLM-OCR | |
|---|---|---|
| Deployment | Hosted API only | Z.ai API or self-hosted with vLLM, SGLang, Ollama or transformers |
| Licence | Proprietary | MIT for model weights; SDK repo and bundled PP-DocLayoutV3 are Apache-2.0 |
| Pricing | PDF $0.005 per page (0-1M), $0.0035 above 1M; images $0.002 each; one-time $19.99 API key activation fee | API $0.03 per million tokens for input and output; self-hosted is free apart from compute |
| Output formats | LaTeX, Markdown, DOCX, HTML, JSON | Markdown, JSON, plain text |
| Benchmarks | Not published | OmniDocBench v1.5 94.62 and olmOCR-Bench 75.2 (authors); ParseBench mean 29.6 (independent) |
| Limits and platform | Images with more than 12 text rows may be billed at the PDF page rate | API caps input at 100 pages, 50 MB per PDF and 10 MB per image; no aarch64 Linux support |
| Known failure modes | Billed monthly in arrears; $19.99 non-refundable fee before the first API key works | Repeated output unless a stop sequence is set; drops table footer rows; loses comma decimal separators in cropped tables |
Sources
- Mathpix billing docs
- Mathpix OCR API
- Hugging Face model card
- Z.ai GLM-OCR docs and pricing
- Issue #240 missing table rows
What each one is
Mathpix Convert
Specialist for math, science papers and handwritten formulas. Outputs LaTeX.
Mathpix Convert detailsTry both on your own file
Run the file once with each model and compare the output.
PDF, PNG, JPG, WebP, HEIC. Up to 20 MB each, 10 files.No file?
Use one, fall back to the other
{"model": "mathpix/convert","fallbacks": ["zai/glm-ocr"],"file_url": "https://example.com/document.pdf"}Questions
More comparisons
Mathpix Convert vs Unlimited OCRMathpix Convert vs PaddleOCR-VLReducto Parse vs GLM-OCRLlamaParse vs GLM-OCRMathpix Convert vs InternVL3.5Mathpix Convert vs EasyOCRMathpix Convert vs NougatGLM-OCR vs Chandra
Mathpix Convert alternativesGLM-OCR alternativesBest OCR models in 2026