Neural networks

MWS TeamStream is aimed at both companies and private clients - freelancers and self-employed individuals.

The dataset will allow developers to train, test, and improve AI-based anti-fraud systems.

The system supports over 70 pre-configured legal entity documents and recognizes data at a speed of 900 pages per minute. The solution will automate retail and business lending processes in banks.

AI Search will allow companies to create their own AI agents based on Yandex search.

It now has a mode for analyzing large amounts of data.

In Direct, artificial intelligence can now adapt headlines and texts to user queries on Search in real time, and also block certain words and phrases in such ads using the "negative keywords" function.

As part of the collaboration, the companies presented a line of AI tools for retail, pharmaceuticals, manufacturing, and other industries.

Starting November 28, Qwen and Kimi with parameter sizes ranging from 235 billion to 1 trillion will be available on the platform.

Alice AI already powers the "Alice AI" neural network, and now business representatives will be able to use this generative model for their tasks and create their own AI applications based on it.

The intelligent assistant provides quick answers to specialists' questions and increases their efficiency.

The system recognizes content created by modern AI models, such as Veo 3 (Google) and Sora 2 (OpenAI), with an accuracy of over 98%.

Yandex B2B Tech has introduced its first specialized AI assistant.

You can give an AI agent a task, and it will independently complete the task of designing and writing code from start to finish.

Sber is currently completing testing of the AI agent and plans to put the first version into commercial operation within the company by the end of the year, and in 2026 other organizations will be able to use the agent.

Sber's platform has received a new GenBI technology (visualization creation on demand).