WORKSPACE

Recognize text locally from images. Results appear below.
📁
Click to choose an image, or drop it here
Paste from clipboard also supported · PNG / JPG / WebP
⚡ OCR_ENGINE::PROCESSING...

HISTORY

Past OCR results with source images. Click image to zoom.

SETTINGS

Configure your OCRMeow experience.

▸ GENERAL

Language
Interface language (Auto detects browser setting)
OCR Model
The underlying AI engine for text recognition

▸ MODEL MANAGEMENT

AI Infrastructure Status
PaddleOCR.js requires both detection and recognition models to function. Models are cached in browser's persistent IndexedDB.
[ DET_CORE_v5 ] 官方源 备用源
MISSING
[ REC_CORE_v5 ] 官方源 备用源
MISSING

▸ APPEARANCE

Capture Theme
Visual theme for the OCR capture overlay and Data Pad
Distortion Intensity
Strength of the liquid distortion effect during OCR. Set to 0 to disable.
0.1
Studio Theme
Theme for this dashboard interface

▸ HISTORY & STORAGE

History Limit
Maximum number of recent OCR results to keep (10 – 9999).
Export downloads a JSON file with text results and source images.