OCRRouter

Qianfan OCR vs Google Cloud Vision OCR

Pick Qianfan OCR when you need structured document parsing you can host yourself: tables, formulas, key information extraction and Markdown for LLM pipelines. Pick Google Cloud Vision OCR when you want a hosted API for plain text from photos, labels and handwritten notes at high volume, with no GPU to run. The main split is structure and control versus a managed text-only API.

Qianfan OCRbaidu/qianfan-ocrGoogle Cloud Vision OCRgoogle/cloud-vision
TypeOCR APIOCR API
Price2 credits / page2 credits / page
Per 1,000 pages$2.00$2.00
Typical speed2.0 s / page1.0 s / page
Best forChinese documentsPhotos and simple scans
InputsPDF, imagesPDF, images
LanguagesChinese, English and more100+ languages
Max pages per request2002,000
Tables✓ YesNo
Handwriting✓ Yes✓ Yes
Math✓ YesNo
Bounding boxesNo✓ Yes
ConfidenceNo✓ Yes
FormsNoNo
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 Qianfan OCR if

  • You need Markdown, JSON, HTML, LaTeX or bounding boxes from one model, including tables and formulas.
  • You want to self-host under Apache-2.0, which allows commercial use, and pay only for compute.
  • You parse documents in many languages: it covers 192 languages, handwriting and scene text, and scores 93.12 on OmniDocBench v1.5 (reported by the authors).

Choose Google Cloud Vision OCR if

  • You only need plain text and bounding polygons, down to page, block, paragraph, word and symbol level.
  • You want a managed API with a free first 1,000 units per month and a price that drops to $0.60 per 1,000 pages above 5M units.
  • You process large batches: async batch accepts up to 2,000 image files per request.

Key differences

Qianfan OCRGoogle Cloud Vision OCR
How you run itOpen weights on Hugging Face, or hosted on the Baidu Qianfan platformHosted API on Google Cloud only, over REST or client libraries
LicenceApache-2.0Proprietary
PriceFree self-hosted (you pay for compute); hosted access via Baidu Qianfan platformFirst 1,000 units per month free, then $1.50 per 1,000 units up to 5M, $0.60 per 1,000 above. One PDF page is one unit
Output formatsMarkdown, JSON, HTML, LaTeX, bounding boxesJSON, plain text, bounding boxes. No Markdown, so you rebuild reading order yourself
Tables and layoutListed as good for tables and formulasNo table or layout structure; Google points to Document AI for forms and tables
BenchmarksOmniDocBench v1.5 overall 93.12, olmOCR-Bench 79.8, OCRBench 880 (authors' numbers)Not published
Known limits1.024 pages per second on an A100 even with W8A8; thinking mode can emit up to 16,384 tokens; not for CPU-only deploymentBilling must be enabled even for the free units; PDF and TIFF need async processing via Cloud Storage; no on-premise deployment

Sources

What each one is

Qianfan OCR

Baidu's OCR-tuned vision model. Strong on Chinese documents and tables.

Qianfan OCR details

Google Cloud Vision OCR

Google Cloud's image OCR with text and document text detection, word boxes and confidence.

Google Cloud Vision OCR 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": "baidu/qianfan-ocr","fallbacks": ["google/cloud-vision"],"file_url": "https://example.com/document.pdf"}

Questions

More comparisons

Qianfan OCR alternativesGoogle Cloud Vision OCR alternativesBest OCR models in 2026