cnstd
Python3 package for Chinese/English Scene Text Detection (STD), Mathematical Formula Detection (MFD), and Layout Analysis, with free pretrained models
Decision gist · record as of 2026-08-14
Yes. Active maintenance, permissive Apache 2.0 license, no known vulnerabilities, and a mature model ecosystem make this a solid choice for text detection tasks. The 20-dependency footprint is typical for ML packages. Install if you need to locate text in images before OCR or analyze document layout; skip if you only need character recognition without localization.AI-flagged interpretation of the facts on this page — verify before relying
Before you install
- Requires Python 3.8 or later.
- First use triggers automatic model download to ~/.cnstd directory; manual download from Hugging Face or Baidu Cloud may be needed if auto-download fails.
- Low installation friction with a pure Python wheel.
License · maintenance · safety
Apache 2.0 (permissive) — Apache 2.0 permissive license allows commercial and private use with minimal restrictions, making it suitable for most production deployments.
last release 2026-07-05 (40 days) · last repo commit 2026-07-05 · 794 stars
0 known vulnerabilities (OSV.dev, 2026-08-14) · 98,858 downloads/mo, #13,055 on PyPI
Alternatives
Verify before relying
pip install cnstd
from cnstd import CnStd
std = CnStd(model_name='multi_PP-OCRv6_det_small')
result = std.detect('path/to/image.jpg')- Inference speed and accuracy metrics for different model variants under various image resolutions and hardware configurations
- Memory footprint and GPU VRAM requirements for each model backend
- Whether layout analysis supports document types beyond the 10 listed categories
What it is and what it does
CnSTD is a Python package for detecting text and other document elements in images. It provides pre-trained models for scene text detection (STD), mathematical formula detection (MFD), and layout analysis, supporting Chinese, English, and multiple other languages. The package includes models trained on standard OCR datasets and external models from PaddleOCR, available in both PyTorch and ONNX formats.
The package is typically used to locate and extract bounding boxes of text regions in images before passing them to an OCR engine for character recognition. It also detects mathematical formulas (inline and isolated) and identifies document layout elements like headers, footers, tables, and figures. Models range from lightweight mobile-optimized variants to larger, more accurate versions, and can run on CPU or GPU.
Use it for
- Locate text regions in document images before feeding them to an OCR system for full text recognition
- Detect and classify mathematical formulas in scientific papers or textbooks for formula extraction pipelines
- Analyze document layout to identify structural elements like titles, tables, and figures for automated document parsing
- Process multilingual documents with language-specific detection models for Chinese, English, or other supported languages
- Deploy lightweight text detection on edge devices using MobileNet or ShuffleNet model variants
Worth the install?
AI-flagged interpretation of the facts on this page. Verify before relying on it.
Yes.
Active maintenance, permissive Apache 2.0 license, no known vulnerabilities, and a mature model ecosystem make this a solid choice for text detection tasks. The 20-dependency footprint is typical for ML packages. Install if you need to locate text in images before OCR or analyze document layout; skip if you only need character recognition without localization.
Install
cnstd on PyPI
Before you install
Low installation friction with a pure Python wheel. Active maintenance with recent release (40 days old) and 794 repository stars. Depends on 20 runtime packages including PyTorch, torchvision, and computer vision libraries—a substantial but standard ML stack.
Requires Python 3.8 or later. First use triggers automatic model download to ~/.cnstd directory; manual download from Hugging Face or Baidu Cloud may be needed if auto-download fails.
License in practice
Apache 2.0 permissive license allows commercial and private use with minimal restrictions, making it suitable for most production deployments.
Quickstart
pip install cnstd
from cnstd import CnStd
std = CnStd(model_name='multi_PP-OCRv6_det_small')
result = std.detect('path/to/image.jpg')Verify before relying
- Inference speed and accuracy metrics for different model variants under various image resolutions and hardware configurations
- Memory footprint and GPU VRAM requirements for each model backend
- Whether layout analysis supports document types beyond the 10 listed categories
Package facts
| License | Apache 2.0 permissive |
| Python support | Supports the current Python release >=3.8 |
| Install friction | Low. Pure-Python wheel |
| Runtime dependencies | 20 packagesclicktqdmpyyamlunidecodetorchtorchvisionnumpyscipypandaspytorch-lightningpillowopencv-pythonshapelypyclippermatplotlibseabornonnxhuggingface-hubultralyticsrapidocr |
| Maintenance | Actively maintained 40 days since the last release |
| Last repo commit | |
| First released | |
| Downloads | 98,858 / month, #13,055 on PyPI 30-day window, as of 2026-08-14 |
| Known vulnerabilities | None known OSV.dev, checked 2026-08-14 |
| Classifiers | Development Status :: 4 - BetaIntended Audience :: DevelopersLicense :: OSI Approved :: Apache Software LicenseOperating System :: OS IndependentProgramming Language :: PythonProgramming Language :: Python :: 3Programming Language :: Python :: 3.10Programming Language :: Python :: 3.8Programming Language :: Python :: 3.9Programming Language :: Python :: ImplementationTopic :: Scientific/Engineering :: Artificial Intelligence |
Evidence: cnstd-1.2.8-py3-none-any.whl
Tags
Let your AI agent find packages like this
Example. Real query, live index.
You found this page by searching. An agent finds it by wishing: SkillFed indexes 14,416 PyPI packages by what they can do, searchable in plain language.
wish › “scene text detection”
- cnstdDetects text and mathematical formulas in images using pre-trained…
- keras-ocrkeras-ocr detects and recognizes text in images using pre-trained…
- cnocrCnOCR recognizes text in images—Chinese (simplified and traditional),…
Give your agent the search over MCP, or paste the wish link into any chat.
More Artificial Intelligence packages
LiteLLM provides a unified Python interface to call 100+ LLM providers (OpenAI, Anthropic, Gemini, Bedrock, Azure, and others) using OpenAI-compatible API format, available as both a Python SDK and a self-hosted AI Gateway proxy server.
Install it if you need to work with multiple LLM providers or want to centralize LLM routing in your organization.
Client library and CLI tool for downloading, uploading, and managing models, datasets, and repositories on the Hugging Face Hub platform.
Install it if you work with Hugging Face Hub models or datasets.
LangChain provides a framework for building agents and LLM-powered applications by composing language models, tools, and memory through a unified API that abstracts over multiple model providers.
hf-xet provides chunk-based deduplication and efficient file transfer for the Hugging Face Hub, enabling faster uploads and downloads of large files with local disk caching.
Tokenizers converts raw text into token sequences for NLP models, with support for training custom vocabularies and using pre-built tokenizers (BPE, WordPiece) optimized for speed via Rust.
Transformers provides a unified framework for loading, fine-tuning, and running state-of-the-art pretrained models across text, vision, audio, video, and multimodal tasks using PyTorch, JAX, or TensorFlow.
Install it if you need to run or train any transformer-based model for NLP, vision, audio, or multimodal tasks.
See also cnocr · rapidocr-onnxruntime · rapidocr · nudenet · paddlex · paddleocr · qrdet · layoutparser · surya-ocr · ocrmac