“Yandex” posted in the public domain a neural network for generating texts

“Yandex” posted in the public domain a neural network for generating texts

The Yandex company has opened public access to the YaLM 100B neural network, which generates and processes texts in Russian and English. To date, this is the largest GPT-like model available to users from all over the world, the company’s press service said. The model is provided under the Apache 2.0 open license and is available on GitHub.

The neural network contains 100 billion parameters – “more than any of the existing models for the Russian language.” With it, you can perform a large number of tasks related to language processing and the creation of texts. For example, create ideas for advertising, product descriptions, and even write poetry.

The model was trained on Yandex supercomputers. During the training, YaLM 100B processed about 2 TB of texts from open resources in English and Russian. The company itself uses the neural network in more than 20 projects. For example, in quick replies in “Search”, the voice assistant “Alice”, for answers in the support service, advertisements and descriptions of sites in snippets


Leave a Comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Recent News

Editor's Pick