Cloud.ru Evolution AI Factory

Main article: PaaS - Platform As A Service

2025: Introduction Cloud.ru Evolution AI Factory

Cloud.ru announced a digital environment for developing applications and agents based on artificial intelligence (AI) in the cloud - Cloud.ru Evolution AI Factory. The solution with the necessary set of tools and services will be available to users in the summer of 2025. This was announced on April 10, 2205 by Yevgeny Kolbin, General Director Cloud.ru.

Cloud.ru Evolution AI Factory is a digital factory of ready-made services. It will combine cloud AI and ML tools for working with generative neural networks and managing the lifecycle of AI-based applications. A versatile environment will provide a wide range of options, from model training to multi-agent system startup. With its help, companies of any scale will be easier to develop AI agents and implement them into business.

The solution will allow you to create your own AI applications using ready-made templates or work with customized AI models. Thanks to a convenient set of technologies and built-in assistants, not only professional developers, but also users without special skills will be able to use the services. Cloud.ru Evolution AI Factory will help you significantly minimize computing infrastructure costs, improve team efficiency, and reduce time to launch new products.

Cloud.ru Evolution AI Factory will include a fully ready infrastructure, services and products for the simple and fast creation of AI applications:

• Foundation Models - fast and convenient access in the Russian cloud to such popular open source AI models as GigaChat, DeepSeek, Qwen and Gemma. The service was created for developers, researchers and companies who need to adapt AI models to their tasks, test hypotheses and conduct experiments - without the need to deploy and optimize complex infrastructure. The service also includes AI Playground - a convenient environment for quick prototyping, hypothesis testing and testing models in practice;
• Evolution ML Inference is a service for running ML models from the Hugging Face library on cloud power with GPUs, including in custom Docker images. It supports vLLM, TGI, Ollama, Diffusers, Transformers execution environments, as well as transformer models, diffusion models, and large language models (LLMs). Inference as a service allows you to divide GPU resources between tasks at the production-ready level of the solution. Thanks to this, users will be able to flexibly and efficiently manage resources in the cloud, reducing costs and speeding up AI models;
• AI Assistants - for creating agent and multi-agent systems with connection to any external systems with MCP (Model Context Protocol). It will help customize the interaction between AI agents and data sources with the ability to use advanced models and proprietary solutions. The service provides quick access to information, automation of routine tasks and support for decision-making with tools such as RAG (retrieval augmented generation);
• Cloud.ru ML Space is a platform for deep learning applications using supercomputers and ML based on the Russian solution. Suitable for the full cycle of ML development in the cloud and on-premium.

Шаблон:Quote 'author=say Evgeny Kolbin, CEO of cloud and AI technology provider Cloud.ru.