Austrian Academy releases Apollo, first AI model for Ancient Greek

September 22, 2026 at 02:00 PM2 min

This digest was compiled by AI from multiple sources — links to the originals are below.

The Austrian Academy of Science released Apollo, the first advanced large language model for Ancient Greek, developed with Mistral and Sail Reply. The model, trained on 600 million historical Greek words, is freely available to academics through a chatbot interface to accelerate papyrus restoration.

Key Facts

Apollo is trained on roughly 600 million historical Greek words from manuscripts, papyri, and inscriptions.
The model was developed by the Austrian Academy of Science in partnership with French AI lab Mistral and technology services firm Sail Reply.
Apollo is freely available to academics through a chatbot interface.
The University of Oxford holds the world's largest ancient papyrus collection.

Model Development

The Austrian Academy of Science released Apollo on Wednesday, developed in partnership with French AI lab Mistral and technology services firm Sail Reply. Apollo is trained on roughly 600 million historical Greek words drawn from manuscripts, papyri, and inscriptions. The model is described as the world's first advanced large language model for Ancient Greek.

Scholarly Impact

Apollo is designed to help scholars more rapidly identify papyrus fragments relevant to their sub-disciplines and promising research avenues. For tattered documents, Apollo fills in missing words or passages with the most statistically likely options, potentially revealing hidden historical details. Stephen Colvin, professor of classics at University College London, noted that very few people in the world possess the specialized knowledge now baked into Apollo. Anna Dolganov, historian and papyrologist at the Austrian Academy of Science, said the model adapts to dialect: 'When it sees Homer, it supplements Homeric Greek. When it sees an inscription in Doric dialect, it uses Doric dialect.'

Expert Reception

Armand D'Angour, professor of classical languages at the University of Oxford, said the tool would 'speed up matters considerably' by suggesting possible words for gaps. Colvin cautioned that Apollo is unlikely to change broad understanding of the ancient world, as many unrestored papyri are mundane documents like personal letters and civil service papers. Dimitris Vlitas, partner at Sail Reply, said unlocking knowledge in this way 'was unthinkable a year ago.'