[Models](/models)/[Text Extraction](/models?category=text_extraction)/PaddlePaddle/paddleocr

PyTorchOCRApache 2.0

# paddleocr

by PaddlePaddle

Ultra-lightweight, production-ready multilingual OCR system

N/Adl/month

12Mparams

[Source](https://github.com/PaddlePaddle/PaddleOCR)[Run it on your own data, free](https://studio.mixpeek.com/namespaces/create?mode=managed&utm_source=homepage&utm_medium=model_page&utm_campaign=mvs_onramp)

Identifiers

Model ID

`PaddlePaddle/paddleocr`

## Deploy paddleocr

Single-tenant

Mixpeek has no managed extractor for this model. On a single-tenant deployment you upload the weights and a custom plugin serves them next to the rest of your pipeline.

[Read the customer-hosted guide](https://mixpeek.com/docs/customer-hosted/overview) [Talk to us about single-tenant](/demo)

## Overview

PaddleOCR is a comprehensive OCR toolkit supporting 80+ languages with extremely lightweight models suitable for both server and mobile deployment. It combines text detection (DB), text direction classification, and text recognition (CRNN) in a unified pipeline.

On Mixpeek, PaddleOCR is the go-to choice for multilingual text extraction and high-throughput OCR processing of documents, images, and video frames.

## Architecture

Three-stage pipeline: (1) DB text detector for localizing text regions, (2) text direction classifier, (3) CRNN-based text recognizer. PP-OCRv4 variant uses knowledge distillation for 4x smaller model with minimal accuracy loss.

## Mixpeek SDK Integration

IngestRetrieve

// No extractor parameter takes a Hugging Face model id (checked against
// GET /v1/discovery/extractors, which returns 13), so paddleocr runs
// on your side and the output is upserted through POST
// /v1/namespaces/{namespace\_id}/documents/upsert. On Enterprise the other
// path is to upload the weights instead: POST /v1/namespaces/{id}/models
// accepts the huggingface format and a custom plugin loads them.
const res = await fetch(
  "https://api.mixpeek.com/v1/namespaces/ns\_your\_namespace/documents/upsert",
  {
    method: "POST",
    headers: {
      Authorization: "Bearer API\_KEY",
      "Content-Type": "application/json",
    },
    body: JSON.stringify({
      collection\_id: "col\_your\_collection",
      documents: \[
        {
          document\_id: "asset-00412",
          // The model produces text, so it lands in payload. Give the
          // collection a text vector index and embed that text to make it
          // searchable rather than only filterable.
          payload: { extracted\_text: modelOutput, source\_key: "archive/2026/asset-00412" },
          vectors: { "text-embedding": embeddingOfModelOutput },
        },
      \],
    }),
  },
);

// Managed alternative, if this exact model is not the requirement:
// universal\_extractor@v1 runs google/gemini-embedding-2
// (3072-d) over a bucket, with no inference of your own.

## Capabilities

*   80+ language support including CJK, Arabic, Devanagari
*   Text detection, recognition, and layout analysis
*   Ultra-lightweight models (< 10MB for mobile)
*   Table recognition and key-value extraction

## Use Cases on Mixpeek

Multilingual document processing across global content libraries

High-throughput OCR for large-scale document digitization

Real-time text extraction from live video feeds

## Benchmarks

Dataset

Metric

Score

Source

ICDAR 2015 (detection)

F1

87.1%

PaddleOCR benchmarks: README

ICDAR 2015 (recognition)

Accuracy

79.4%

PaddleOCR benchmarks: README

## Performance

Input Sizevariable (auto-resize)

GPU Latency~12ms / page (A100)

CPU Latency~85ms / page

GPU Throughput~83 pages/sec (A100)

GPU Memory~0.6 GB

Includes detection + recognition pipeline

## Common Pipeline Companions

[

BAAI/bge-large-en-v1.5

Embed extracted text for search

](/model/BAAI/bge-large-en-v1.5)[

microsoft/layoutlmv3-base

Layout analysis for document structure

](/model/microsoft/layoutlmv3-base)

## Explore on Mixpeek

[

More Text Extraction models

Compare alternatives in this category

](/models?category=text_extraction)[

Best Document AI Platforms

Hand-picked tools & platforms compared

](/curated-lists/best-document-ai-platforms)[

OCR & Document AI Internals

Deep-dive technical guide

](/guides/ocr-document-ai-internals)[

Feature Extractors

See how Mixpeek runs models as extractors

](/extractors)[

Mixpeek Vector Store

Store & search embeddings at scale

](/mvs)[

Pricing

Usage-based pricing for pipelines

](/pricing)[

All Curated Lists

Compare models, APIs & infrastructure

](/curated-lists)

### Specification

FrameworkPyTorch

OrganizationPaddlePaddle

FeatureOCR

Outputtext + bbox

Modalitiesvideo, image, document

RetrieverText-in-Image

Parameters12M

LicenseApache 2.0

Downloads/moN/A

### Build a pipeline with paddleocr

Add this model to a processing pipeline alongside other extractors. Combine with retrieval stages for end-to-end search.

[Run it on your own data, free](https://studio.mixpeek.com/namespaces/create?mode=managed&utm_source=homepage&utm_medium=model_page&utm_campaign=mvs_onramp)

### Alternative Models

[

microsoft/trocr-large-printed

OCR

](/model/microsoft/trocr-large-printed)[

zai-org/GLM-OCR

OCR

](/model/zai-org/GLM-OCR)[

lightonai/LightOnOCR-2-1B

OCR

](/model/lightonai/LightOnOCR-2-1B)[

deepseek-ai/DeepSeek-OCR-2

OCR

](/model/deepseek-ai/DeepSeek-OCR-2)

### Related in Text Extraction

[

microsoft/codebert-base

Code Extraction

](/model/microsoft/codebert-base)[

Salesforce/codet5p-110m-embedding

Code Extraction

](/model/Salesforce/codet5p-110m-embedding)[

nomic-ai/nomic-embed-code

Code Extraction

](/model/nomic-ai/nomic-embed-code)[

Salesforce/SFR-Embedding-Code-2B\_R

Code Extraction

](/model/Salesforce/SFR-Embedding-Code-2B_R)