---
title: "Read text from a picture · OpenVibe.Vision"
canonical: "https://openvibe.vision/ocr"
updated: null
---

# Read text from a picture · OpenVibe.Vision

HTML: [Read text from a picture · OpenVibe.Vision](https://openvibe.vision/ocr)

Local first

# Read text from a picture

Tesseract 5 runs on this host. Nothing is sent to a model unless you are signed in, tick the fallback, and the local pass finds no usable text. The reading is stored as text; the image is not.

 Upload an image …or a public image URL 

 Language If the local reader finds no text, ask OpenVibe.AI (needs a sign-in) 

[Ask a question instead](/ask)

## No image yet

Upload a PNG, JPEG, WebP, GIF, BMP, TIFF or AVIF and press Read locally. The words come back in reading order.