RSS
Логотип
Баннер в шапке 1
Баннер в шапке 2
Project

Gazprom Transgaz Tomsk is introducing the GigaChat neural network

Customers: Gazprom Transgaz Tomsk

Contractors: Sberbank
Product: Sberbank: GigaChat (GigaChat)

Project date: 2025/10

2025: Implementation of GigaChat

Together with Sberbank "" Gazprom Transgaz Tomsk implements neuronet GigaChat for business to automate the verification of as-built documentation for construction linear objects. This repeatedly speeds up the acceptance of objects and minimizes errors in the coordination of documents. Sberbank announced this on November 21, 2025.

The construction of linear facilities, especially by a wide range of contractors, is often associated with a large amount of manual work to check closing documents: certificates of work performed and inspected hidden works and other forms. As of November 2025, specialists manually verify data from hundreds of documents with project documentation, which slows down the commissioning of facilities.

{{quote 'author=said Andrei Dmitriyev, Head of CIB Sberbank's Customer Transformation Directorate. | This solution based on the GigaChat neural network automatically recognizes documents, including tables, using the modality for working with images and compares the volumes of work according to different acts with the volumes indicated in the pick list, and at the same time checks spelling and punctuation. The high speed and quality of the result are provided by AI agents for self-analysis of context, choice of actions and solving complex problems. With our partner Gazprom Transgaz Tomsk, we are conducting a pilot project to automate the verification of as-built documentation. The first results showed that our solution saves time and eliminates errors, increasing the quality and speed of construction control,}}

Шаблон:Quote 'author=noted Vladislav Borodin, General Director of Gazprom Transgaz Tomsk LLC.

In the future, the massive introduction of an AI assistant into linear construction throughout the country will dramatically speed up Stroycontrol and allow much faster commissioning of facilities. Acceptance specialists will be able to devote more time to problematic issues, and artificial intelligence will take over the entire routine.