Hieroglyphs AI identifies the Gardiner code of an Egyptian hieroglyph in an image using a deep neural network.

Draw a box around the hieroglyph of interest: drag with the mouse, or with your finger on a phone or tablet. For successful recognition, the sign should be roughly in the middle of the box and fully inside it (not cut off).

The neural network was trained on about 145,000 hand-labelled hieroglyphs of 709 classes from stelae, temple walls, tombs and sign lists. It shows the five most likely Gardiner codes with their probabilities.

On monuments it has never seen, the correct code is the first answer in about 89% of cases and among the top five in about 97%.

Questions or suggestions? Write to info@hieroglyphs.ai.

How to cite. Sulimov, A. (2026). Hieroglyphs AI: automatic Gardiner code recognition for Egyptian hieroglyphs [Technical note]. Zenodo. https://doi.org/10.5281/zenodo.23691717

@misc{hieroglyphsai2026,
  author    = {Sulimov, Alexander},
  title     = {Hieroglyphs AI: automatic Gardiner code recognition for Egyptian hieroglyphs},
  year      = {2026},
  publisher = {Zenodo},
  note      = {Technical note},
  doi       = {10.5281/zenodo.23691717},
  url       = {https://hieroglyphs.ai}
}