OCRRouter

GLM-OCR vs Chandra

Pick GLM-OCR for high-volume PDF to Markdown, academic papers with formulas and schema-based field extraction, where its $0.03 per million token API price matters most. Pick Chandra for multilingual scans and documents with tables, forms and math, where it scores 85.8 on olmOCR-bench. Watch the licence: Chandra needs a paid commercial licence for self-hosting above $2M in funding or revenue, while GLM-OCR weights are MIT.

GLM-OCRzai/glm-ocrChandradatalab/chandra
TypeOpen sourceOpen source
Price6 credits / page2 credits / page
Per 1,000 pages$6.00$2.00
Typical speed1.3 s / page2.0 s / page
Best forLow-cost general documentsComplex layouts and handwriting
InputsPDF, imagesPDF, images
Languages100+ languages40+ languages
Max pages per request500500
Tables✓ Yes✓ Yes
HandwritingNo✓ Yes
Math✓ Yes✓ Yes
Bounding boxes✓ Yes✓ Yes
Confidence✓ YesNo
FormsNo✓ Yes
Multilingual✓ Yes✓ Yes
On OCRRouterAvailableAvailable

Prices are what you pay on OCRRouter. Speed is the provider's typical figure, not a benchmark we ran.

Which to choose

Choose GLM-OCR if

  • The API costs $0.03 per million tokens for input and output, and the weights are MIT licensed with no revenue limit.
  • It is a 0.9B model that processed 1.86 PDF pages per second on a single replica in the authors' test.
  • It can extract fields into a user-defined JSON schema, which suits forms, and it scores 94.62 on OmniDocBench v1.5 per its model card.

Choose Chandra if

  • It scores 85.8 on olmOCR-bench (Chandra 2), above dots.ocr 1.5 at 83.9 and olmOCR 2 at 82.4 in the authors' table.
  • It averages 77.8% across 43 languages, versus 67.6% for Gemini 2.5 Flash.
  • It returns Markdown, HTML and JSON plus layout chunks with bounding boxes and labels for each block.

Key differences

GLM-OCRChandra
LicenceMIT for the weights; the SDK repo and bundled PP-DocLayoutV3 are Apache-2.0Apache-2.0 code; modified OpenRAIL-M weights, free for research, personal use and companies under $2M funding or revenue
Hosted API price$0.03 per million tokens for input and output$1 (fast), $3 (balanced) or $7.50 (accurate) per 1,000 pages; free tier gives $20 per month with a work email
Model size and hardware0.9B parameters; cannot run on aarch64 LinuxSingle NVIDIA GPU via vLLM; benchmarks on one H100 80GB, with presets for A100, L40S and RTX 4090/5090
Output formatsMarkdown, JSON and plain text; schema-based field extractionMarkdown, HTML, JSON and bounding boxes
Benchmarks94.62 on OmniDocBench v1.5 and 75.2 on olmOCR-bench (authors); 29.6 mean on ParseBench (independent)85.8 on olmOCR-bench, 77.8% on a 43-language benchmark and 72.7% on a 90-language benchmark (authors)
Languages and scansNot publishedWeak on low-resource scripts: 43.2% Urdu, 51.1% Sanskrit; Old Scans category scores 51.1 on olmOCR-bench
Known failuresRepeats output unless a stop sequence is set (issue #220); drops table footer rows (#240); loses comma decimal separators in cropped tables (#241)Token-loop hallucinations on 5-6 of about 60 Russian and Kazakh scans (issue #51); Datalab says raw output hallucinates on about 6 in 10 of its hardest pages
API limits100 pages, 50 MB per PDF and 10 MB per imageNot published

Sources

What each one is

GLM-OCR

Small open-source OCR model from Z.ai with layout regions. MIT licensed.

GLM-OCR details

Chandra

Datalab's OCR model that keeps layout and handles forms, tables and handwriting.

Chandra details

Try 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": "zai/glm-ocr","fallbacks": ["datalab/chandra"],"file_url": "https://example.com/document.pdf"}

Questions

More comparisons

GLM-OCR alternativesChandra alternativesBest OCR models in 2026