Skip to content
Agentmart
Submit a tool

Image to Text (OCR)

Extracts text from an image using OCR that runs in your browser.

Artifact by iib0011 (OmniTools)

IndexedMITDesign & imagesNo sign-upRuns in your browser

Use it

Opens omnitools.app in a new tab. Nothing to install.

You provide
Image with text
You get
Recognised text
Account
No sign-up needed
AI
No AI: a plain tool
Your input
Processed in your browser, not uploaded (stated by the tool or its repo)
Price
Free to use
Made with
Open-source web app
Source
Open source
Link check
Answered on Oct 31, 2026

About

Image to Text (OCR) is an open-source web app in the design media category (ocr). It opens in your browser, so there is nothing to install. You provide image with text and get recognised text. Your input is processed on your device and not uploaded. It needs no account. It is a plain tool: no AI model is involved. The source code is public on GitHub (iib0011/omni-tools).

Category: Design & images · ocr

No reviews yet. Used it? Sign in with an established GitHub account to rate it. How reviews work

More design & images tools

See all design & images tools →

Details

Auth
No auth
License
MIT
First seen
Oct 31, 2026
Last checked
Oct 31, 2026
Source
GitHub

More from iib0011 (OmniTools)

Cite this page

Image to Text (OCR): free online tool. Agent Mart. https://agentmart.dev/artifacts/image-to-text-ocr (last updated 2026-10-31)