Cloud provider MWS Cloud took second place in the updated GPU Cloud 2026 rating, demonstrating a transition from providing computing power to creating full-fledged ML ecosystems.
What Happened
According to the CNews Market rating, MWS Cloud entered the top three leaders of the GPU Cloud 2026 market, sharing second place with Turbo Oblako. The provider's infrastructure is built on VMware and includes more than 30 types of processors and 13 types of GPU accelerators. To support artificial intelligence tasks, the company is developing specialized services: MWS GPT Models Hub for working with LLMs, Inference Valve, a machine learning platform, and the U-Speech Analytics speech analytics tool.
Context
The modern Russian GPU infrastructure market is transforming: the period of shortage and focus on model training is being replaced by the stage of industrial exploitation (inference). The development of services such as LLM Hub and ready-made ML platforms indicates market maturity, where not only "raw" computing power but also integrated technological stacks are in demand.
Why It Matters for the Industry
For the industry, this means increased competition among local providers through the offering of high-level services rather than just hardware rental. The shift in focus toward inference tools promotes the standardization of "ML-as-a-Service" models and the formation of a full-fledged AI-as-a-Service market in Russia.
Why It Matters for Users
Developers and companies gain the ability to instantly scale AI projects and accelerate time-to-market. The availability of ready-to-use server configurations and specialized ML environments allows them to focus on algorithm development, eliminating the complexities of independent procurement and setup of expensive equipment.
Sources
Author
Look at AI, Editorial Office